Documentation / RelNotes / 2.12.0.txton commit Git 2.12-rc1 (5588dbf)
   1Git 2.12 Release Notes
   2======================
   3
   4Backward compatibility notes.
   5
   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).
  12
  13 * The historical argument order "git merge <msg> HEAD <commit>..."
  14   has been deprecated for quite some time, and will be removed in a
  15   future release.
  16
  17 * An ancient script "git relink" has been removed.
  18
  19
  20Updates since v2.11
  21-------------------
  22
  23UI, Workflows & Features
  24
  25 * Various updates to "git p4".
  26
  27 * "git p4" didn't interact with the internal of .git directory
  28   correctly in the modern "git-worktree"-enabled world.
  29
  30 * "git branch --list" and friends learned "--ignore-case" option to
  31   optionally sort branches and tags case insensitively.
  32
  33 * In addition to %(subject), %(body), "log --pretty=format:..."
  34   learned a new placeholder %(trailers).
  35
  36 * "git rebase" learned "--quit" option, which allows a user to
  37   remove the metadata left by an earlier "git rebase" that was
  38   manually aborted without using "git rebase --abort".
  39
  40 * "git clone --reference $there --recurse-submodules $super" has been
  41   taught to guess repositories usable as references for submodules of
  42   $super that are embedded in $there while making a clone of the
  43   superproject borrow objects from $there; extend the mechanism to
  44   also allow submodules of these submodules to borrow repositories
  45   embedded in these clones of the submodules embedded in the clone of
  46   the superproject.
  47
  48 * Porcelain scripts written in Perl are getting internationalized.
  49
  50 * "git merge --continue" has been added as a synonym to "git commit"
  51   to conclude a merge that has stopped due to conflicts.
  52
  53 * Finer-grained control of what protocols are allowed for transports
  54   during clone/fetch/push have been enabled via a new configuration
  55   mechanism.
  56
  57 * "git shortlog" learned "--committer" option to group commits by
  58   committer, instead of author.
  59
  60 * GitLFS integration with "git p4" has been updated.
  61
  62 * The isatty() emulation for Windows has been updated to eradicate
  63   the previous hack that depended on internals of (older) MSVC
  64   runtime.
  65
  66 * Some platforms no longer understand "latin-1" that is still seen in
  67   the wild in e-mail headers; replace them with "iso-8859-1" that is
  68   more widely known when conversion fails from/to it.
  69
  70 * "git grep" has been taught to optionally recurse into submodules.
  71
  72 * "git rm" used to refuse to remove a submodule when it has its own
  73   git repository embedded in its working tree.  It learned to move
  74   the repository away to $GIT_DIR/modules/ of the superproject
  75   instead, and allow the submodule to be deleted (as long as there
  76   will be no loss of local modifications, that is).
  77
  78 * A recent updates to "git p4" was not usable for older p4 but it
  79   could be made to work with minimum changes.  Do so.
  80
  81 * "git diff" learned diff.interHunkContext configuration variable
  82   that gives the default value for its --inter-hunk-context option.
  83
  84 * The prereleaseSuffix feature of version comparison that is used in
  85   "git tag -l" did not correctly when two or more prereleases for the
  86   same release were present (e.g. when 2.0, 2.0-beta1, and 2.0-beta2
  87   are there and the code needs to compare 2.0-beta1 and 2.0-beta2).
  88
  89 * "git submodule push" learned "--recurse-submodules=only option to
  90   push submodules out without pushing the top-level superproject.
  91
  92 * "git tag" and "git verify-tag" learned to put GPG verification
  93   status in their "--format=<placeholders>" output format.
  94
  95 * An ancient repository conversion tool left in contrib/ has been
  96   removed.
  97
  98 * "git show-ref HEAD" used with "--verify" because the user is not
  99   interested in seeing refs/remotes/origin/HEAD, and used with
 100   "--head" because the user does not want HEAD to be filtered out,
 101   i.e. "git show-ref --head --verify HEAD", did not work as expected.
 102
 103 * "git submodule add" used to be confused and refused to add a
 104   locally created repository; users can now use "--force" option
 105   to add them.
 106   (merge 619acfc78c sb/submodule-add-force later to maint).
 107
 108 * Some people feel the default set of colors used by "git log --graph"
 109   rather limiting.  A mechanism to customize the set of colors has
 110   been introduced.
 111
 112 * "git read-tree" and its underlying unpack_trees() machinery learned
 113   to report problematic paths prefixed with the --super-prefix option.
 114
 115 * When a submodule "A", which has another submodule "B" nested within
 116   it, is "absorbed" into the top-level superproject, the inner
 117   submodule "B" used to be left in a strange state.  The logic to
 118   adjust the .git pointers in these submodules has been corrected.
 119
 120 * The user can specify a custom update method that is run when
 121   "submodule update" updates an already checked out submodule.  This
 122   was ignored when checking the submodule out for the first time and
 123   we instead always just checked out the commit that is bound to the
 124   path in the superproject's index.
 125
 126 * The command line completion (in contrib/) learned that
 127   "git diff --submodule=" can take "diff" as a recently added option.
 128
 129 * The "core.logAllRefUpdates" that used to be boolean has been
 130   enhanced to take 'always' as well, to record ref updates to refs
 131   other than the ones that are expected to be updated (i.e. branches,
 132   remote-tracking branches and notes).
 133
 134
 135Performance, Internal Implementation, Development Support etc.
 136
 137 * Commands that operate on a log message and add lines to the trailer
 138   blocks, such as "format-patch -s", "cherry-pick (-x|-s)", and
 139   "commit -s", have been taught to use the logic of and share the
 140   code with "git interpret-trailer".
 141
 142 * The default Travis-CI configuration specifies newer P4 and GitLFS.
 143
 144 * The "fast hash" that had disastrous performance issues in some
 145   corner cases has been retired from the internal diff.
 146
 147 * The character width table has been updated to match Unicode 9.0
 148
 149 * Update the procedure to generate "tags" for developer support.
 150
 151 * The codeflow of setting NOATIME and CLOEXEC on file descriptors Git
 152   opens has been simplified.
 153
 154 * "git diff" and its family had two experimental heuristics to shift
 155   the contents of a hunk to make the patch easier to read.  One of
 156   them turns out to be better than the other, so leave only the
 157   "--indent-heuristic" option and remove the other one.
 158
 159 * A new submodule helper "git submodule embedgitdirs" to make it
 160   easier to move embedded .git/ directory for submodules in a
 161   superproject to .git/modules/ (and point the latter with the former
 162   that is turned into a "gitdir:" file) has been added.
 163
 164 * "git push \\server\share\dir" has recently regressed and then
 165   fixed.  A test has retroactively been added for this breakage.
 166
 167 * Build updates for Cygwin.
 168
 169 * The implementation of "real_path()" was to go there with chdir(2)
 170   and call getcwd(3), but this obviously wouldn't be usable in a
 171   threaded environment.  Rewrite it to manually resolve relative
 172   paths including symbolic links in path components.
 173
 174 * Adjust documentation to help AsciiDoctor render better while not
 175   breaking the rendering done by AsciiDoc.
 176
 177 * The sequencer machinery has been further enhanced so that a later
 178   set of patches can start using it to reimplement "rebase -i".
 179
 180 * Update the definition of the MacOSX test environment used by
 181   TravisCI.
 182
 183 * Rewrite a scripted porcelain "git difftool" in C.
 184
 185 * "make -C t failed" will now run only the tests that failed in the
 186   previous run.  This is usable only when prove is not use, and gives
 187   a useless error message when run after "make clean", but otherwise
 188   is serviceable.
 189
 190 * "uchar [40]" to "struct object_id" conversion continues.
 191
 192
 193Also contains various documentation updates and code clean-ups.
 194
 195Fixes since v2.10
 196-----------------
 197
 198Unless otherwise noted, all the fixes since v2.9 in the maintenance
 199track are contained in this release (see the maintenance releases'
 200notes for details).
 201
 202 * We often decide if a session is interactive by checking if the
 203   standard I/O streams are connected to a TTY, but isatty() that
 204   comes with Windows incorrectly returned true if it is used on NUL
 205   (i.e. an equivalent to /dev/null).  This has been fixed.
 206
 207 * "git svn" did not work well with path components that are "0", and
 208   some configuration variable it uses were not documented.
 209
 210 * "git rev-parse --symbolic" failed with a more recent notation like
 211   "HEAD^-1" and "HEAD^!".
 212
 213 * An empty directory in a working tree that can simply be nuked used
 214   to interfere while merging or cherry-picking a change to create a
 215   submodule directory there, which has been fixed..
 216
 217 * The code in "git push" to compute if any commit being pushed in the
 218   superproject binds a commit in a submodule that hasn't been pushed
 219   out was overly inefficient, making it unusable even for a small
 220   project that does not have any submodule but have a reasonable
 221   number of refs.
 222
 223 * "git push --dry-run --recurse-submodule=on-demand" wasn't
 224   "--dry-run" in the submodules.
 225
 226 * The output from "git worktree list" was made in readdir() order,
 227   and was unstable.
 228
 229 * mergetool.<tool>.trustExitCode configuration variable did not apply
 230   to built-in tools, but now it does.
 231
 232 * "git p4" LFS support was broken when LFS stores an empty blob.
 233
 234 * A corner case in merge-recursive regression that crept in
 235   during 2.10 development cycle has been fixed.
 236
 237 * Transport with dumb http can be fooled into following foreign URLs
 238   that the end user does not intend to, especially with the server
 239   side redirects and http-alternates mechanism, which can lead to
 240   security issues.  Tighten the redirection and make it more obvious
 241   to the end user when it happens.
 242
 243 * Update the error messages from the dumb-http client when it fails
 244   to obtain loose objects; we used to give sensible error message
 245   only upon 404 but we now forbid unexpected redirects that needs to
 246   be reported with something sensible.
 247
 248 * When diff.renames configuration is on (and with Git 2.9 and later,
 249   it is enabled by default, which made it worse), "git stash"
 250   misbehaved if a file is removed and another file with a very
 251   similar content is added.
 252
 253 * "git diff --no-index" did not take "--no-abbrev" option.
 254
 255 * "git difftool --dir-diff" had a minor regression when started from
 256   a subdirectory, which has been fixed.
 257
 258 * "git commit --allow-empty --only" (no pathspec) with dirty index
 259   ought to be an acceptable way to create a new commit that does not
 260   change any paths, but it was forbidden, perhaps because nobody
 261   needed it so far.
 262
 263 * Git 2.11 had a minor regression in "merge --ff-only" that competed
 264   with another process that simultanously attempted to update the
 265   index. We used to explain what went wrong with an error message,
 266   but the new code silently failed.  The error message has been
 267   resurrected.
 268
 269 * A pathname that begins with "//" or "\\" on Windows is special but
 270   path normalization logic was unaware of it.
 271
 272 * "git pull --rebase", when there is no new commits on our side since
 273   we forked from the upstream, should be able to fast-forward without
 274   invoking "git rebase", but it didn't.
 275
 276 * The way to specify hotkeys to "xxdiff" that is used by "git
 277   mergetool" has been modernized to match recent versions of xxdiff.
 278
 279 * Unlike "git am --abort", "git cherry-pick --abort" moved HEAD back
 280   to where cherry-pick started while picking multiple changes, when
 281   the cherry-pick stopped to ask for help from the user, and the user
 282   did "git reset --hard" to a different commit in order to re-attempt
 283   the operation.
 284
 285 * Code cleanup in shallow boundary computation.
 286
 287 * A recent update to receive-pack to make it easier to drop garbage
 288   objects made it clear that GIT_ALTERNATE_OBJECT_DIRECTORIES cannot
 289   have a pathname with a colon in it (no surprise!), and this in turn
 290   made it impossible to push into a repository at such a path.  This
 291   has been fixed by introducing a quoting mechanism used when
 292   appending such a path to the colon-separated list.
 293
 294 * The function usage_msg_opt() has been updated to say "fatal:"
 295   before the custom message programs give, when they want to die
 296   with a message about wrong command line options followed by the
 297   standard usage string.
 298
 299 * "git index-pack --stdin" needs an access to an existing repository,
 300   but "git index-pack file.pack" to generate an .idx file that
 301   corresponds to a packfile does not.
 302
 303 * Fix for NDEBUG builds.
 304
 305 * A lazy "git push" without refspec did not internally use a fully
 306   specified refspec to perform 'current', 'simple', or 'upstream'
 307   push, causing unnecessary "ambiguous ref" errors.
 308
 309 * "git p4" misbehaved when swapping a directory and a symbolic link.
 310
 311 * Even though an fix was attempted in Git 2.9.3 days, but running
 312   "git difftool --dir-diff" from a subdirectory never worked. This
 313   has been fixed.
 314
 315 * "git p4" that tracks multile p4 paths imported a single changelist
 316   that touches files in these multiple paths as one commit, followed
 317   by many empty commits.  This has been fixed.
 318
 319 * A potential but unlikely buffer overflow in Windows port has been
 320   fixed.
 321
 322 * When the http server gives an incomplete response to a smart-http
 323   rpc call, it could lead to client waiting for a full response that
 324   will never come.  Teach the client side to notice this condition
 325   and abort the transfer.
 326
 327 * Compression setting for producing packfiles were spread across
 328   three codepaths, one of which did not honor any configuration.
 329   Unify these so that all of them honor core.compression and
 330   pack.compression variables the same way.
 331
 332 * "git fast-import" sometimes mishandled while rebalancing notes
 333   tree, which has been fixed.
 334
 335 * Recent update to the default abbreviation length that auto-scales
 336   lacked documentation update, which has been corrected.
 337
 338 * Leakage of lockfiles in the config subsystem has been fixed.
 339
 340 * It is natural that "git gc --auto" may not attempt to pack
 341   everything into a single pack, and there is no point in warning
 342   when the user has configured the system to use the pack bitmap,
 343   leading to disabling further "gc".
 344
 345 * "git archive" did not read the standard configuration files, and
 346   failed to notice a file that is marked as binary via the userdiff
 347   driver configuration.
 348
 349 * "git blame --porcelain" misidentified the "previous" <commit, path>
 350   pair (aka "source") when contents came from two or more files.
 351
 352 * "git rebase -i" with a recent update started showing an incorrect
 353   count when squashing more than 10 commits.
 354
 355 * "git <cmd> @{push}" on a detached HEAD used to segfault; it has
 356   been corrected to error out with a message.
 357
 358 * Running "git add a/b" when "a" is a submodule correctly errored
 359   out, but without a meaningful error message.
 360   (merge 2d81c48fa7 sb/pathspec-errors later to maint).
 361
 362 * Typing ^C to pager, which usually does not kill it, killed Git and
 363   took the pager down as a collateral damage in certain process-tree
 364   structure.  This has been fixed.
 365
 366 * "git mergetool" without any pathspec on the command line that is
 367   run from a subdirectory became no-op in Git v2.11 by mistake, which
 368   has been fixed.
 369
 370 * Retire long unused/unmaintained gitview from the contrib/ area.
 371   (merge 3120925c25 sb/remove-gitview later to maint).
 372
 373 * Tighten a test to avoid mistaking an extended ERE regexp engine as
 374   a PRE regexp engine.
 375
 376 * An error message with an ASCII control character like '\r' in it
 377   can alter the message to hide its early part, which is problematic
 378   when a remote side gives such an error message that the local side
 379   will relay with a "remote: " prefix.
 380   (merge f290089879 jk/vreport-sanitize later to maint).
 381
 382 * "git fsck" inspects loose objects more carefully now.
 383   (merge cce044df7f jk/loose-object-fsck later to maint).
 384
 385 * A crashing bug introduced in v2.11 timeframe has been found (it is
 386   triggerable only in fast-import) and fixed.
 387   (merge abd5a00268 jk/clear-delta-base-cache-fix later to maint).
 388
 389 * With an anticipatory tweak for remotes defined in ~/.gitconfig
 390   (e.g. "remote.origin.prune" set to true, even though there may or
 391   may not actually be "origin" remote defined in a particular Git
 392   repository), "git remote rename" and other commands misinterpreted
 393   and behaved as if such a non-existing remote actually existed.
 394   (merge e459b073fb js/remote-rename-with-half-configured-remote later to maint).
 395
 396 * A few codepaths had to rely on a global variable when sorting
 397   elements of an array because sort(3) API does not allow extra data
 398   to be passed to the comparison function.  Use qsort_s() when
 399   natively available, and a fallback implementation of it when not,
 400   to eliminate the need, which is a prerequisite for making the
 401   codepath reentrant.
 402
 403 * "git fsck --connectivity-check" was not working at all.
 404   (merge a2b22854bd jk/fsck-connectivity-check-fix later to maint).
 405
 406 * After starting "git rebase -i", which first opens the user's editor
 407   to edit the series of patches to apply, but before saving the
 408   contents of that file, "git status" failed to show the current
 409   state (i.e. you are in an interactive rebase session, but you have
 410   applied no steps yet) correctly.
 411   (merge df9ded4984 js/status-pre-rebase-i later to maint).
 412
 413 * Test tweak for FreeBSD where /usr/bin/unzip is unsuitable to run
 414   our tests but /usr/local/bin/unzip is usable.
 415   (merge d98b2c5fce js/unzip-in-usr-bin-workaround later to maint).
 416
 417 * "git p4" did not work well with multiple git-p4.mapUser entries on
 418   Windows.
 419   (merge c3c2b05776 gv/mingw-p4-mapuser later to maint).
 420
 421 * "git help" enumerates executable files in $PATH; the implementation
 422   of "is this file executable?" on Windows has been optimized.
 423   (merge c755015f79 hv/mingw-help-is-executable later to maint).
 424
 425 * Test tweaks for those who have default ACL in their git source tree
 426   that interfere with the umask test.
 427   (merge d549d21307 mm/reset-facl-before-umask-test later to maint).
 428
 429 * Names of the various hook scripts must be spelled exactly, but on
 430   Windows, an .exe binary must be named with .exe suffix; notice
 431   $GIT_DIR/hooks/<hookname>.exe as a valid <hookname> hook.
 432   (merge 235be51fbe js/mingw-hooks-with-exe-suffix later to maint).
 433
 434 * Asciidoctor, an alternative reimplementation of AsciiDoc, still
 435   needs some changes to work with documents meant to be formatted
 436   with AsciiDoc.  "make USE_ASCIIDOCTOR=YesPlease" to use it out of
 437   the box to document our pages is getting closer to reality.
 438
 439 * Correct command line completion (in contrib/) on "git svn"
 440   (merge 2cbad17642 ew/complete-svn-authorship-options later to maint).
 441
 442 * Incorrect usage help message for "git worktree prune" has been fixed.
 443   (merge 2488dcab22 ps/worktree-prune-help-fix later to maint).
 444
 445 * Adjust a perf test to new world order where commands that do
 446   require a repository are really strict about having a repository.
 447   (merge c86000c1a7 rs/p5302-create-repositories-before-tests later to maint).
 448
 449 * "git log --graph" did not work well with "--name-only", even though
 450   other forms of "diff" output were handled correctly.
 451   (merge f5022b5fed jk/log-graph-name-only later to maint).
 452
 453 * Other minor doc, test and build updates and code cleanups.
 454   (merge f2627d9b19 sb/submodule-config-cleanup later to maint).
 455   (merge 384f1a167b sb/unpack-trees-cleanup later to maint).
 456   (merge 874444b704 rh/diff-orderfile-doc later to maint).
 457   (merge eafd5d9483 cw/doc-sign-off later to maint).
 458   (merge 0aaad415bc rs/absolute-pathdup later to maint).
 459   (merge 4432dd6b5b rs/receive-pack-cleanup later to maint).
 460   (merge 540a398e9c sg/mailmap-self later to maint).
 461   (merge 209df269a6 nd/rev-list-all-includes-HEAD-doc later to maint).