{"thread":{"id":"45915","subject":"[ANNOUNCE] Git v2.13.0","startedAt":"2017-05-10T00:04:33Z","lastAt":"2017-05-10T00:04:33Z","messageCount":1,"participants":["Junio C Hamano"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"319242","messageId":"xmqqa86lzifl.fsf@gitster.mtv.corp.google.com","threadId":"45915","inReplyTo":null,"subject":"[ANNOUNCE] Git v2.13.0","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2017-05-10T00:03:58Z","receivedAt":"2017-05-10T00:04:33Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"The latest feature release Git v2.13.0 is now available at the\nusual places.  It is comprised of 729 non-merge commits since\nv2.12.0, contributed by 65 people, 15 of which are new faces.\n\nThis release also contains the security patch in v2.12.3 and\nothers to fix CVE-2017-8386.\n\nThe tarballs are found at:\n\n    https://www.kernel.org/pub/software/scm/git/\n\nThe following public repositories all have a copy of the 'v2.13.0'\ntag and the 'master' branch that the tag points at:\n\n  url = https://kernel.googlesource.com/pub/scm/git/git\n  url = git://repo.or.cz/alt-git.git\n  url = git://git.sourceforge.jp/gitroot/git-core/git.git\n  url = git://git-core.git.sourceforge.net/gitroot/git-core/git-core\n  url = https://github.com/gitster/git\n\nNew contributors whose contributions weren't in v2.12.0 are as follows.\nWelcome to the Git development community!\n\n  Allan Xavier, Andreas Heiduk, Devin J. Pohly, Devin Lehmacher,\n  Hiroshi Shirosaki, Johan Hovold, Maxim Moseychuk, Mostyn\n  Bramley-Moore, Prathamesh Chavan, Quentin Pradet, René Genz,\n  Segev Finer, Sergey Ryazanov, Stephen Hicks, and Valery Tolstov.\n\nReturning contributors who helped this release are as follows.\nThanks for your continued support.\n\n  Ævar Arnfjörð Bjarmason, Alexander Shopov, Alex Henrie,\n  Brandon Williams, brian m. carlson, Christian Couder, Cornelius\n  Weig, David Aguilar, David Turner, Eric Wong, Giuseppe Bilotta,\n  Jacob Keller, Jean-Noel Avila, Jeff Hostetler, Jeff King,\n  Jiang Xin, Johannes Schindelin, Jonathan Nieder, Jonathan Tan,\n  Jordi Mas, Junio C Hamano, Karthik Nayak, Kevin Willford,\n  Kyle Meyer, Lars Schneider, Linus Torvalds, Luke Diamand, Matt\n  McCutchen, Michael Haggerty, Michael J Gruber, Michael Rappazzo,\n  Mike Hommey, Nguyễn Thái Ngọc Duy, Patrick Steinhardt,\n  Peter Krefting, Ralf Thielow, Ramsay Jones, Ray Chen, René\n  Scharfe, Ross Lagerwall, Santiago Torres, Sebastian Schuberth,\n  Simon Ruderich, Stefan Beller, SZEDER Gábor, Thomas Gummerer,\n  Torsten Bögershausen, Trần Ngọc Quân, Vasco Almeida,\n  and Vegard Nossum.\n\n----------------------------------------------------------------\n\nGit 2.13 Release Notes\n======================\n\nBackward compatibility notes.\n\n * Use of an empty string as a pathspec element that is used for\n   'everything matches' is still warned and Git asks users to use a\n   more explicit '.' for that instead.  The hope is that existing\n   users will not mind this change, and eventually the warning can be\n   turned into a hard error, upgrading the deprecation into removal of\n   this (mis)feature.  That is not scheduled to happen in the upcoming\n   release (yet).\n\n * The historical argument order \"git merge <msg> HEAD <commit>...\"\n   has been deprecated for quite some time, and is now removed.\n\n * The default location \"~/.git-credential-cache/socket\" for the\n   socket used to communicate with the credential-cache daemon has\n   been moved to \"~/.cache/git/credential/socket\".\n\n * Git now avoids blindly falling back to \".git\" when the setup\n   sequence said we are _not_ in Git repository.  A corner case that\n   happens to work right now may be broken by a call to die(\"BUG\").\n   We've tried hard to locate such cases and fixed them, but there\n   might still be cases that need to be addressed--bug reports are\n   greatly appreciated.\n\n\nUpdates since v2.12\n-------------------\n\nUI, Workflows & Features\n\n * \"git describe\" and \"git name-rev\" have been taught to take more\n   than one refname patterns to restrict the set of refs to base their\n   naming output on, and also learned to take negative patterns to\n   name refs not to be used for naming via their \"--exclude\" option.\n\n * Deletion of a branch \"foo/bar\" could remove .git/refs/heads/foo\n   once there no longer is any other branch whose name begins with\n   \"foo/\", but we didn't do so so far.  Now we do.\n\n * When \"git merge\" detects a path that is renamed in one history\n   while the other history deleted (or modified) it, it now reports\n   both paths to help the user understand what is going on in the two\n   histories being merged.\n\n * The <url> part in \"http.<url>.<variable>\" configuration variable\n   can now be spelled with '*' that serves as wildcard.\n   E.g. \"http.https://*.example.com.proxy\" can be used to specify the\n   proxy used for https://a.example.com, https://b.example.com, etc.,\n   i.e. any host in the example.com domain.\n\n * \"git tag\" did not leave useful message when adding a new entry to\n   reflog; this was left unnoticed for a long time because refs/tags/*\n   doesn't keep reflog by default.\n\n * The \"negative\" pathspec feature was somewhat more cumbersome to use\n   than necessary in that its short-hand used \"!\" which needed to be\n   escaped from shells, and it required \"exclude from what?\" specified.\n\n * The command line options for ssh invocation needs to be tweaked for\n   some implementations of SSH (e.g. PuTTY plink wants \"-P <port>\"\n   while OpenSSH wants \"-p <port>\" to specify port to connect to), and\n   the variant was guessed when GIT_SSH environment variable is used\n   to specify it.  The logic to guess now applies to the command\n   specified by the newer GIT_SSH_COMMAND and also core.sshcommand\n   configuration variable, and comes with an escape hatch for users to\n   deal with misdetected cases.\n\n * The \"--git-path\", \"--git-common-dir\", and \"--shared-index-path\"\n   options of \"git rev-parse\" did not produce usable output.  They are\n   now updated to show the path to the correct file, relative to where\n   the caller is.\n\n * \"git diff -W\" has been taught to handle the case where a new\n   function is added at the end of the file better.\n\n * \"git update-ref -d\" and other operations to delete references did\n   not leave any entry in HEAD's reflog when the reference being\n   deleted was the current branch.  This is not a problem in practice\n   because you do not want to delete the branch you are currently on,\n   but caused renaming of the current branch to something else not to\n   be logged in a useful way.\n\n * \"Cc:\" on the trailer part does not have to conform to RFC strictly,\n   unlike in the e-mail header.  \"git send-email\" has been updated to\n   ignore anything after '>' when picking addresses, to allow non-address\n   cruft like \" # stable 4.4\" after the address.\n\n * When \"git submodule init\" decides that the submodule in the working\n   tree is its upstream, it now gives a warning as it is not a very\n   common setup.\n\n * \"git stash push\" takes a pathspec so that the local changes can be\n   stashed away only partially.\n\n * Documentation for \"git ls-files\" did not refer to core.quotePath.\n\n * The experimental \"split index\" feature has gained a few\n   configuration variables to make it easier to use.\n\n * From a working tree of a repository, a new option of \"rev-parse\"\n   lets you ask if the repository is used as a submodule of another\n   project, and where the root level of the working tree of that\n   project (i.e. your superproject) is.\n\n * The pathspec mechanism learned to further limit the paths that\n   match the pattern to those that have specified attributes attached\n   via the gitattributes mechanism.\n\n * Our source code has used the SHA1_HEADER cpp macro after \"#include\"\n   in the C code to switch among the SHA-1 implementations. Instead,\n   list the exact header file names and switch among implementations\n   using \"#ifdef BLK_SHA1/#include \"block-sha1/sha1.h\"/.../#endif\";\n   this helps some IDE tools.\n\n * The start-up sequence of \"git\" needs to figure out some configured\n   settings before it finds and set itself up in the location of the\n   repository and was quite messy due to its \"chicken-and-egg\" nature.\n   The code has been restructured.\n\n * The command line prompt (in contrib/) learned a new 'tag' style\n   that can be specified with GIT_PS1_DESCRIBE_STYLE, to describe a\n   detached HEAD with \"git describe --tags\".\n\n * The configuration file learned a new \"includeIf.<condition>.path\"\n   that includes the contents of the given path only when the\n   condition holds.  This allows you to say \"include this work-related\n   bit only in the repositories under my ~/work/ directory\".\n\n * Recent update to \"rebase -i\" started showing a message that is not\n   a warning with \"warning:\" prefix by mistake.  This has been fixed.\n\n * Recently we started passing the \"--push-options\" through the\n   external remote helper interface; now the \"smart HTTP\" remote\n   helper understands what to do with the passed information.\n\n * \"git describe --dirty\" dies when it cannot be determined if the\n   state in the working tree matches that of HEAD (e.g. broken\n   repository or broken submodule).  The command learned a new option\n   \"git describe --broken\" to give \"$name-broken\" (where $name is the\n   description of HEAD) in such a case.\n\n * \"git checkout\" is taught the \"--recurse-submodules\" option.\n\n * Recent enhancement to \"git stash push\" command to support pathspec\n   to allow only a subset of working tree changes to be stashed away\n   was found to be too chatty and exposed the internal implementation\n   detail (e.g. when it uses reset to match the index to HEAD before\n   doing other things, output from reset seeped out).  These, and\n   other chattyness has been fixed.\n\n * \"git merge <message> HEAD <commit>\" syntax that has been deprecated\n   since October 2007 has been removed.\n\n * The refs completion for large number of refs has been sped up,\n   partly by giving up disambiguating ambiguous refs and partly by\n   eliminating most of the shell processing between 'git for-each-ref'\n   and 'ls-remote' and Bash's completion facility.\n\n * On many keyboards, typing \"@{\" involves holding down SHIFT key and\n   one can easily end up with \"@{Up...\" when typing \"@{upstream}\".  As\n   the upstream/push keywords do not appear anywhere else in the syntax,\n   we can safely accept them case insensitively without introducing\n   ambiguity or confusion to solve this.\n\n * \"git tag/branch/for-each-ref\" family of commands long allowed to\n   filter the refs by \"--contains X\" (show only the refs that are\n   descendants of X), \"--merged X\" (show only the refs that are\n   ancestors of X), \"--no-merged X\" (show only the refs that are not\n   ancestors of X).  One curious omission, \"--no-contains X\" (show\n   only the refs that are not descendants of X) has been added to\n   them.\n\n * The default behaviour of \"git log\" in an interactive session has\n   been changed to enable \"--decorate\".\n\n * The output from \"git status --short\" has been extended to show\n   various kinds of dirtyness in submodules differently; instead of to\n   \"M\" for modified, 'm' and '?' can be shown to signal changes only\n   to the working tree of the submodule but not the commit that is\n   checked out.\n\n * Allow the http.postbuffer configuration variable to be set to a\n   size that can be expressed in size_t, which can be larger than\n   ulong on some platforms.\n\n * \"git rebase\" learns \"--signoff\" option.\n\n * The completion script (in contrib/) learned to complete \"git push\n   --delete b<TAB>\" to complete branch name to be deleted.\n\n * \"git worktree add --lock\" allows to lock a worktree immediately\n   after it's created. This helps prevent a race between \"git worktree\n   add; git worktree lock\" and \"git worktree prune\".\n\n * Completion for \"git checkout <branch>\" that auto-creates the branch\n   out of a remote tracking branch can now be disabled, as this\n   completion often gets in the way when completing to checkout an\n   existing local branch that happens to share the same prefix with\n   bunch of remote tracking branches.\n\n\nPerformance, Internal Implementation, Development Support etc.\n\n * The code to list branches in \"git branch\" has been consolidated\n   with the more generic ref-filter API.\n\n * Resource usage while enumerating refs from alternate object store\n   has been optimized to help receiving end of \"push\" that hosts a\n   repository with many \"forks\".\n\n * The gitattributes machinery is being taught to work better in a\n   multi-threaded environment.\n\n * \"git rebase -i\" starts using the recently updated \"sequencer\" code.\n\n * Code and design clean-up for the refs API.\n\n * The preload-index code has been taught not to bother with the index\n   entries that are paths that are not checked out by \"sparse checkout\".\n\n * Some warning() messages from \"git clean\" were updated to show the\n   errno from failed system calls.\n\n * The \"parse_config_key()\" API function has been cleaned up.\n\n * A test that creates a confusing branch whose name is HEAD has been\n   corrected not to do so.\n\n * The code that parses header fields in the commit object has been\n   updated for (micro)performance and code hygiene.\n\n * An helper function to make it easier to append the result from\n   real_path() to a strbuf has been added.\n\n * Reduce authentication round-trip over HTTP when the server supports\n   just a single authentication method.  This also improves the\n   behaviour when Git is misconfigured to enable http.emptyAuth\n   against a server that does not authenticate without a username\n   (i.e. not using Kerberos etc., which makes http.emptyAuth\n   pointless).\n\n * Windows port wants to use OpenSSL's implementation of SHA-1\n   routines, so let them.\n\n * The t/perf performance test suite was not prepared to test not so\n   old versions of Git, but now it covers versions of Git that are not\n   so ancient.\n\n * Add 32-bit Linux variant to the set of platforms to be tested with\n   Travis CI.\n\n * \"git branch --list\" takes the \"--abbrev\" and \"--no-abbrev\" options\n   to control the output of the object name in its \"-v\"(erbose)\n   output, but a recent update started ignoring them; fix it before\n   the breakage reaches to any released version.\n\n * Picking two versions of Git and running tests to make sure the\n   older one and the newer one interoperate happily has now become\n   possible.\n\n * \"git tag --contains\" used to (ab)use the object bits to keep track\n   of the state of object reachability without clearing them after\n   use; this has been cleaned up and made to use the newer commit-slab\n   facility.\n\n * The \"debug\" helper used in the test framework learned to run\n   a command under \"gdb\" interactively.\n\n * The \"detect attempt to create collisions\" variant of SHA-1\n   implementation by Marc Stevens (CWI) and Dan Shumow (Microsoft)\n   has been integrated and made the default.\n\n * The test framework learned to detect unterminated here documents.\n\n * The name-hash used for detecting paths that are different only in\n   cases (which matter on case insensitive filesystems) has been\n   optimized to take advantage of multi-threading when it makes sense.\n\n * An earlier version of sha1dc/sha1.c that was merged to 'master'\n   compiled incorrectly on Windows, which has been fixed.\n\n * \"what URL do we want to update this submodule?\" and \"are we\n   interested in this submodule?\" are split into two distinct\n   concepts, and then the way used to express the latter got extended,\n   paving a way to make it easier to manage a project with many\n   submodules and make it possible to later extend use of multiple\n   worktrees for a project with submodules.\n\n * Some debugging output from \"git describe\" were marked for l10n,\n   but some weren't.  Mark missing ones for l10n.\n\n * Define a new task in .travis.yml that triggers a test session on\n   Windows run elsewhere.\n\n * Conversion from uchar[20] to struct object_id continues.\n\n * The \"submodule\" specific field in the ref_store structure is\n   replaced with a more generic \"gitdir\" that can later be used also\n   when dealing with ref_store that represents the set of refs visible\n   from the other worktrees.\n\n * The string-list API used a custom reallocation strategy that was\n   very inefficient, instead of using the usual ALLOC_GROW() macro,\n   which has been fixed.\n   (merge 950a234cbd jh/string-list-micro-optim later to maint).\n\n * In a 2- and 3-way merge of trees, more than one source trees often\n   end up sharing an identical subtree; optimize by not reading the\n   same tree multiple times in such a case.\n   (merge d12a8cf0af jh/unpack-trees-micro-optim later to maint).\n\n * The index file has a trailing SHA-1 checksum to detect file\n   corruption, and historically we checked it every time the index\n   file is used.  Omit the validation during normal use, and instead\n   verify only in \"git fsck\".\n\n * Having a git command on the upstream side of a pipe in a test\n   script will hide the exit status from the command, which may cause\n   us to fail to notice a breakage; rewrite tests in a script to avoid\n   this issue.\n\n * Travis CI learns to run coccicheck.\n\n * \"git checkout\" that handles a lot of paths has been optimized by\n   reducing the number of unnecessary checks of paths in the\n   has_dir_name() function.\n\n * The internals of the refs API around the cached refs has been\n   streamlined.\n\n * Output from perf tests have been updated to align their titles.\n\nAlso contains various documentation updates and code clean-ups.\n\n\nFixes since v2.12\n-----------------\n\nUnless otherwise noted, all the fixes since v2.12 in the maintenance\ntrack are contained in this release (see the maintenance releases'\nnotes for details).\n\n * \"git repack --depth=<n>\" for a long time busted the specified depth\n   when reusing delta from existing packs.  This has been corrected.\n\n * The code to parse the command line \"git grep <patterns>... <rev>\n   [[--] <pathspec>...]\" has been cleaned up, and a handful of bugs\n   have been fixed (e.g. we used to check \"--\" if it is a rev).\n\n * \"git ls-remote\" and \"git archive --remote\" are designed to work\n   without being in a directory under Git's control.  However, recent\n   updates revealed that we randomly look into a directory called\n   .git/ without actually doing necessary set-up when working in a\n   repository.  Stop doing so.\n\n * \"git show-branch\" expected there were only very short branch names\n   in the repository and used a fixed-length buffer to hold them\n   without checking for overflow.\n\n * A caller of tempfile API that uses stdio interface to write to\n   files may ignore errors while writing, which is detected when\n   tempfile is closed (with a call to ferror()).  By that time, the\n   original errno that may have told us what went wrong is likely to\n   be long gone and was overwritten by an irrelevant value.\n   close_tempfile() now resets errno to EIO to make errno at least\n   predictable.\n\n * \"git remote rm X\", when a branch has remote X configured as the\n   value of its branch.*.remote, tried to remove branch.*.remote and\n   branch.*.merge and failed if either is unset.\n\n * A \"gc.log\" file left by a backgrounded \"gc --auto\" disables further\n   automatic gc; it has been taught to run at least once a day (by\n   default) by ignoring a stale \"gc.log\" file that is too old.\n\n * The code to parse \"git -c VAR=VAL cmd\" and set configuration\n   variable for the duration of cmd had two small bugs, which have\n   been fixed.\n\n * user.email that consists of only cruft chars should consistently\n   error out, but didn't.\n\n * \"git upload-pack\", which is a counter-part of \"git fetch\", did not\n   report a request for a ref that was not advertised as invalid.\n   This is generally not a problem (because \"git fetch\" will stop\n   before making such a request), but is the right thing to do.\n\n * A leak in a codepath to read from a packed object in (rare) cases\n   has been plugged.\n\n * When a redirected http transport gets an error during the\n   redirected request, we ignored the error we got from the server,\n   and ended up giving a not-so-useful error message.\n\n * The patch subcommand of \"git add -i\" was meant to have paths\n   selection prompt just like other subcommand, unlike \"git add -p\"\n   directly jumps to hunk selection.  Recently, this was broken and\n   \"add -i\" lost the paths selection dialog, but it now has been\n   fixed.\n\n * Git v2.12 was shipped with an embarrassing breakage where various\n   operations that verify paths given from the user stopped dying when\n   seeing an issue, and instead later triggering segfault.\n\n * There is no need for Python only to give a few messages to the\n   standard error stream, but we somehow did.\n\n * The code to parse \"git log -L...\" command line was buggy when there\n   are many ranges specified with -L; overrun of the allocated buffer\n   has been fixed.\n\n * The command-line parsing of \"git log -L\" copied internal data\n   structures using incorrect size on ILP32 systems.\n\n * \"git diff --quiet\" relies on the size field in diff_filespec to be\n   correctly populated, but diff_populate_filespec() helper function\n   made an incorrect short-cut when asked only to populate the size\n   field for paths that need to go through convert_to_git() (e.g. CRLF\n   conversion).\n\n * A few tests were run conditionally under (rare) conditions where\n   they cannot be run (like running cvs tests under 'root' account).\n\n * \"git branch @\" created refs/heads/@ as a branch, and in general the\n   code that handled @{-1} and @{upstream} was a bit too loose in\n   disambiguating.\n\n * \"git fetch\" that requests a commit by object name, when the other\n   side does not allow such an request, failed without much\n   explanation.\n\n * \"git filter-branch --prune-empty\" drops a single-parent commit that\n   becomes a no-op, but did not drop a root commit whose tree is empty.\n\n * Recent versions of Git treats http alternates (used in dumb http\n   transport) just like HTTP redirects and requires the client to\n   enable following it, due to security concerns.  But we forgot to\n   give a warning when we decide not to honor the alternates.\n\n * \"git push\" had a handful of codepaths that could lead to a deadlock\n   when unexpected error happened, which has been fixed.\n\n * \"Dumb http\" transport used to misparse a nonsense http-alternates\n   response, which has been fixed.\n\n * \"git add -p <pathspec>\" unnecessarily expanded the pathspec to a\n   list of individual files that matches the pathspec by running \"git\n   ls-files <pathspec>\", before feeding it to \"git diff-index\" to see\n   which paths have changes, because historically the pathspec\n   language supported by \"diff-index\" was weaker.  These days they are\n   equivalent and there is no reason to internally expand it.  This\n   helps both performance and avoids command line argument limit on\n   some platforms.\n   (merge 7288e12cce jk/add-i-use-pathspecs later to maint).\n\n * \"git status --porcelain\" is supposed to give a stable output, but a\n   few strings were left as translatable by mistake.\n\n * \"git revert -m 0 $merge_commit\" complained that reverting a merge\n   needs to say relative to which parent the reversion needs to\n   happen, as if \"-m 0\" weren't given.  The correct diagnosis is that\n   \"-m 0\" does not refer to the first parent (\"-m 1\" does).  This has\n   been fixed.\n\n * Code to read submodule.<name>.ignore config did not state the\n   variable name correctly when giving an error message diagnosing\n   misconfiguration.\n\n * Fix for NO_PTHREADS build.\n\n * Fix for potential segv introduced in v2.11.0 and later (also\n   v2.10.2) to \"git log --pickaxe-regex -S\".\n\n * A few unterminated here documents in tests were fixed, which in\n   turn revealed incorrect expectations the tests make. These tests\n   have been updated.\n\n * Fix for NO_PTHREADS option.\n   (merge 2225e1ea20 bw/grep-recurse-submodules later to maint).\n\n * Git now avoids blindly falling back to \".git\" when the setup\n   sequence said we are _not_ in Git repository.  A corner case that\n   happens to work right now may be broken by a call to die(\"BUG\").\n   (merge b1ef400eec jk/no-looking-at-dotgit-outside-repo-final later to maint).\n\n * A few commands that recently learned the \"--recurse-submodule\"\n   option misbehaved when started from a subdirectory of the\n   superproject.\n   (merge b2dfeb7c00 bw/recurse-submodules-relative-fix later to maint).\n\n * FreeBSD implementation of getcwd(3) behaved differently when an\n   intermediate directory is unreadable/unsearchable depending on the\n   length of the buffer provided, which our strbuf_getcwd() was not\n   aware of.  strbuf_getcwd() has been taught to cope with it better.\n   (merge a54e938e5b rs/freebsd-getcwd-workaround later to maint).\n\n * A recent update to \"rebase -i\" stopped running hooks for the \"git\n   commit\" command during \"reword\" action, which has been fixed.\n\n * Removing an entry from a notes tree and then looking another note\n   entry from the resulting tree using the internal notes API\n   functions did not work as expected.  No in-tree users of the API\n   has such access pattern, but it still is worth fixing.\n\n * \"git receive-pack\" could have been forced to die by attempting\n   allocate an unreasonably large amount of memory with a crafted push\n   certificate; this has been fixed.\n   (merge f2214dede9 bc/push-cert-receive-fix later to maint).\n\n * Update error handling for codepath that deals with corrupt loose\n   objects.\n   (merge 51054177b3 jk/loose-object-info-report-error later to maint).\n\n * \"git diff --submodule=diff\" learned to work better in a project\n   with a submodule that in turn has its own submodules.\n   (merge 17b254cda6 sb/show-diff-for-submodule-in-diff-fix later to maint).\n\n * Update the build dependency so that an update to /usr/bin/perl\n   etc. result in recomputation of perl.mak file.\n   (merge c59c4939c2 ab/regen-perl-mak-with-different-perl later to maint).\n\n * \"git push --recurse-submodules --push-option=<string>\" learned to\n   propagate the push option recursively down to pushes in submodules.\n\n * If a patch e-mail had its first paragraph after an in-body header\n   indented (even after a blank line after the in-body header line),\n   the indented line was mistook as a continuation of the in-body\n   header.  This has been fixed.\n   (merge fd1062e52e lt/mailinfo-in-body-header-continuation later to maint).\n\n * Clean up fallouts from recent tightening of the set-up sequence,\n   where Git barfs when repository information is accessed without\n   first ensuring that it was started in a repository.\n   (merge bccb22cbb1 jk/no-looking-at-dotgit-outside-repo later to maint).\n\n * \"git p4\" used \"name-rev HEAD\" when it wants to learn what branch is\n   checked out; it should use \"symbolic-ref HEAD\".\n   (merge eff451101d ld/p4-current-branch-fix later to maint).\n\n * \"http.proxy\" set to an empty string is used to disable the usage of\n   proxy.  We broke this early last year.\n   (merge ae51d91105 sr/http-proxy-configuration-fix later to maint).\n\n * $GIT_DIR may in some cases be normalized with all symlinks resolved\n   while \"gitdir\" path expansion in the pattern does not receive the\n   same treatment, leading to incorrect mismatch.  This has been fixed.\n\n * \"git submodule\" script does not work well with strange pathnames.\n   Protect it from a path with slashes in them, at least.\n\n * \"git fetch-pack\" was not prepared to accept ERR packet that the\n   upload-pack can send with a human-readable error message.  It\n   showed the packet contents with ERR prefix, so there was no data\n   loss, but it was redundant to say \"ERR\" in an error message.\n   (merge 8e2c7bef03 jt/fetch-pack-error-reporting later to maint).\n\n * \"ls-files --recurse-submodules\" did not quite work well in a\n   project with nested submodules.\n\n * gethostname(2) may not NUL terminate the buffer if hostname does\n   not fit; unfortunately there is no easy way to see if our buffer\n   was too small, but at least this will make sure we will not end up\n   using garbage past the end of the buffer.\n   (merge 5781a9a270 dt/xgethostname-nul-termination later to maint).\n\n * A recent update broke \"git add -p ../foo\" from a subdirectory.\n\n * While handy, \"git_path()\" is a dangerous function to use as a\n   callsite that uses it safely one day can be broken by changes\n   to other code that calls it.  Reduction of its use continues.\n   (merge 16d2676c9e jk/war-on-git-path later to maint).\n\n * The split-index code configuration code used an unsafe git_path()\n   function without copying its result out.\n\n * Many stale HTTP(s) links have been updated in our documentation.\n   (merge 613416f0be jk/update-links-in-docs later to maint).\n\n * \"git-shell\" rejects a request to serve a repository whose name\n   begins with a dash, which makes it no longer possible to get it\n   confused into spawning service programs like \"git-upload-pack\" with\n   an option like \"--help\", which in turn would spawn an interactive\n   pager, instead of working with the repository user asked to access\n   (i.e. the one whose name is \"--help\").\n\n * Other minor doc, test and build updates and code cleanups.\n   (merge df2a6e38b7 jk/pager-in-use later to maint).\n   (merge 75ec4a6cb0 ab/branch-list-doc later to maint).\n   (merge 3e5b36c637 sg/skip-prefix-in-prettify-refname later to maint).\n   (merge 2c5e2865cc jk/fast-import-cleanup later to maint).\n   (merge 4473060bc2 ab/test-readme-updates later to maint).\n   (merge 48a96972fd ab/doc-submitting later to maint).\n   (merge f5c2bc2b96 jk/make-coccicheck-detect-errors later to maint).\n   (merge c105f563d1 cc/untracked later to maint).\n   (merge 8668976b53 jc/unused-symbols later to maint).\n   (merge fba275dc93 jc/bs-t-is-not-a-tab-for-sed later to maint).\n   (merge be6ed145de mm/ls-files-s-doc later to maint).\n   (merge 60b091c679 qp/bisect-docfix later to maint).\n   (merge 47242cd103 ah/diff-files-ours-theirs-doc later to maint).\n   (merge 35ad44cbd8 sb/submodule-rm-absorb later to maint).\n   (merge 0301f1fd92 va/i18n-perl-scripts later to maint).\n   (merge 733e064d98 vn/revision-shorthand-for-side-branch-log later to maint).\n   (merge 85999743e7 tb/doc-eol-normalization later to maint).\n   (merge 0747fb49fd jk/loose-object-fsck later to maint).\n   (merge d8f4481c4f jk/quarantine-received-objects later to maint).\n   (merge 7ba1ceef95 xy/format-patch-base later to maint).\n   (merge fa1912c89a rs/misc-cppcheck-fixes later to maint).\n   (merge f17d642d3b ab/push-cas-doc-n-test later to maint).\n   (merge 61e282425a ss/gitmodules-ignore-doc later to maint).\n   (merge 8d3047cd5b ss/submodule-shallow-doc later to maint).\n   (merge 1f9e18b772 jk/prio-queue-avoid-swap-with-self later to maint).\n   (merge 627fde1025 jk/submodule-init-segv-fix later to maint).\n   (merge d395745d81 rg/doc-pull-typofix later to maint).\n   (merge 01e60a9a22 rg/doc-submittingpatches-wordfix later to maint).\n   (merge 501d3cd7b8 sr/hooks-cwd-doc later to maint).\n\n----------------------------------------------------------------\n\nChanges since v2.12.0 are as follows:\n\nAlex Henrie (1):\n      log: if --decorate is not given, default to --decorate=auto\n\nAlexander Shopov (2):\n      l10n: bg.po: Updated Bulgarian translation (3201t)\n      l10n: bg.po: Updated Bulgarian translation (3195t)\n\nAllan Xavier (1):\n      line-log.c: prevent crash during union of too many ranges\n\nAndreas Heiduk (2):\n      Documentation: improve description for core.quotePath\n      diff-files: document --ours etc.\n\nBrandon Williams (37):\n      attr: pass struct attr_check to collect_some_attrs\n      attr: use hashmap for attribute dictionary\n      attr: eliminate global check_all_attr array\n      attr: remove maybe-real, maybe-macro from git_attr\n      attr: tighten const correctness with git_attr and match_attr\n      attr: store attribute stack in attr_check structure\n      attr: push the bare repo check into read_attr()\n      attr: reformat git_attr_set_direction() function\n      pathspec: allow querying for attributes\n      pathspec: allow escaped query values\n      submodule--helper: add is-active subcommand\n      submodule status: use submodule--helper is-active\n      submodule sync: skip work for inactive submodules\n      submodule sync: use submodule--helper is-active\n      submodule--helper clone: check for configured submodules using helper\n      grep: fix help text typo\n      setup: allow for prefix to be passed to git commands\n      grep: fix bug when recursing with relative pathspec\n      ls-files: fix typo in variable name\n      ls-files: fix bug when recursing with relative pathspec\n      grep: set default output method\n      submodule: decouple url and submodule interest\n      submodule init: initialize active submodules\n      clone: teach --recurse-submodules to optionally take a pathspec\n      submodule--helper init: set submodule.<name>.active\n      submodule add: respect submodule.active and submodule.<name>.active\n      grep: fix builds with with no thread support\n      send-pack: send push options correctly in stateless-rpc case\n      remote-curl: allow push options\n      sequencer: fix missing newline\n      push: unmark a local variable as static\n      push: propagate push-options with --recurse-submodules\n      remote: expose parse_push_refspec function\n      submodule--helper: add push-check subcommand\n      push: propagate remote and refspec with --recurse-submodules\n      pathspec: fix segfault in clear_pathspec\n      submodule: prevent backslash expantion in submodule names\n\nChristian Couder (25):\n      config: mark an error message up for translation\n      t1700: change here document style\n      config: add git_config_get_split_index()\n      split-index: add {add,remove}_split_index() functions\n      read-cache: add and then use tweak_split_index()\n      update-index: warn in case of split-index incoherency\n      t1700: add tests for core.splitIndex\n      Documentation/config: add information for core.splitIndex\n      Documentation/git-update-index: talk about core.splitIndex config var\n      config: add git_config_get_max_percent_split_change()\n      read-cache: regenerate shared index if necessary\n      t1700: add tests for splitIndex.maxPercentChange\n      Documentation/config: add splitIndex.maxPercentChange\n      sha1_file: make check_and_freshen_file() non static\n      read-cache: touch shared index files when used\n      config: add git_config_get_expiry() from gc.c\n      read-cache: unlink old sharedindex files\n      t1700: test shared index file expiration\n      read-cache: refactor read_index_from()\n      read-cache: use freshen_shared_index() in read_index_from()\n      Documentation/config: add splitIndex.sharedIndexExpire\n      Documentation/git-update-index: explain splitIndex.*\n      update-index: fix xgetcwd() related memory leak\n      p0004: make perf test executable\n      read-cache: avoid using git_path() in freshen_shared_index()\n\nCornelius Weig (1):\n      tag: generate useful reflog message\n\nDavid Aguilar (3):\n      t7800: remove whitespace before redirect\n      t7800: cleanup cruft left behind by tests\n      difftool: handle modified symlinks in dir-diff mode\n\nDavid Turner (3):\n      gc: ignore old gc.log files\n      http.postbuffer: allow full range of ssize_t values\n      xgethostname: handle long hostnames\n\nDevin J. Pohly (4):\n      t7003: ensure --prune-empty can prune root commit\n      t7003: ensure --prune-empty removes entire branch when applicable\n      filter-branch: fix --prune-empty on parentless commits\n      p7000: add test for filter-branch with --prune-empty\n\nDevin Lehmacher (3):\n      path.c: add xdg_cache_home\n      credential-cache: use XDG_CACHE_HOME for socket\n      credential-cache: add tests for XDG functionality\n\nEric Wong (3):\n      README: create HTTP/HTTPS links from URLs in Markdown\n      http: inform about alternates-as-redirects behavior\n      http: release strbuf on disabled alternates\n\nGiuseppe Bilotta (4):\n      builtin/am: honor --signoff also when --rebasing\n      builtin/am: fold am_signoff() into am_append_signoff()\n      doc: trivial typo in git-format-patch.txt\n      rebase: pass --[no-]signoff option to git am\n\nHiroshi Shirosaki (1):\n      git svn: fix authentication with 'branch'\n\nJacob Keller (7):\n      doc: add documentation for OPT_STRING_LIST\n      name-rev: extend --refs to accept multiple patterns\n      name-rev: add support to exclude refs by pattern match\n      describe: teach --match to accept multiple patterns\n      describe: teach describe negative pattern matches\n      ls-files: fix recurse-submodules with nested submodules\n      ls-files: fix path used when recursing into submodules\n\nJean-Noel Avila (3):\n      i18n: remove i18n from tag reflog message\n      l10n: fr.po v2.13 round 1\n      l10n: fr.po v2.13 rnd 2\n\nJeff Hostetler (19):\n      mingw: use OpenSSL's SHA-1 routines\n      preload-index: avoid lstat for skip-worktree items\n      name-hash: specify initial size for istate.dir_hash table\n      hashmap: allow memihash computation to be continued\n      hashmap: add disallow_rehash setting\n      hashmap: document memihash_cont, hashmap_disallow_rehash api\n      name-hash: perf improvement for lazy_init_name_hash\n      name-hash: add test-lazy-init-name-hash\n      name-hash: add perf test for lazy_init_name_hash\n      test-online-cpus: helper to return cpu count\n      read-cache: force_verify_index_checksum\n      string-list: use ALLOC_GROW macro when reallocing string_list\n      read-cache: add strcmp_offset function\n      unpack-trees: avoid duplicate ODB lookups during checkout\n      p0006-read-tree-checkout: perf test to time read-tree\n      read-cache: speed up add_index_entry during checkout\n      read-cache: speed up has_dir_name (part 1)\n      read-cache: speed up has_dir_name (part 2)\n      t1450: avoid use of \"sed\" on the index, which is a binary file\n\nJeff King (132):\n      setup_git_env: avoid blind fall-back to \".git\"\n      pack-objects: enforce --depth limit in reused deltas\n      pack-objects: convert recursion to iteration in break_delta_chain()\n      for_each_alternate_ref: handle failure from real_pathdup()\n      for_each_alternate_ref: stop trimming trailing slashes\n      for_each_alternate_ref: use strbuf for path allocation\n      for_each_alternate_ref: pass name/oid instead of ref struct\n      for_each_alternate_ref: replace transport code with for-each-ref\n      fetch-pack: cache results of for_each_alternate_ref\n      add oidset API\n      receive-pack: use oidset to de-duplicate .have lines\n      receive-pack: fix misleading namespace/.have comment\n      receive-pack: treat namespace .have lines like alternates\n      receive-pack: avoid duplicates between our refs and alternates\n      grep: move thread initialization a little lower\n      grep: re-order rev-parsing loop\n      grep: fix \"--\" rev/pathspec disambiguation\n      grep: avoid resolving revision names in --no-index case\n      grep: do not diagnose misspelt revs with --no-index\n      show-branch: drop head_len variable\n      show-branch: store resolved head in heap buffer\n      remote: avoid reading $GIT_DIR config in non-repo\n      grep: treat revs the same for --untracked as for --no-index\n      show-branch: use skip_prefix to drop magic numbers\n      tempfile: set errno to a known value before calling ferror()\n      http: restrict auth methods to what the server advertises\n      ident: mark error messages for translation\n      ident: handle NULL email when complaining of empty name\n      ident: reject all-crud ident name\n      ident: do not ignore empty config name/email\n      parse_config_key: use skip_prefix instead of starts_with\n      parse_config_key: allow matching single-level config\n      parse_hide_refs_config: tell parse_config_key we don't want a subsection\n      http: add an \"auto\" mode for http.emptyauth\n      t6300: avoid creating refs/heads/HEAD\n      add--interactive: fix missing file prompt for patch mode with \"-i\"\n      interpret_branch_name(): handle auto-namelen for @{-1}\n      interpret_branch_name: move docstring to header file\n      strbuf_branchname: drop return value\n      strbuf_branchname: add docstring\n      interpret_branch_name: allow callers to restrict expansions\n      t3204: test git-branch @-expansion corner cases\n      branch: restrict @-expansions when deleting\n      strbuf_check_ref_format(): expand only local branches\n      checkout: restrict @-expansions when finding branch\n      t/perf: use $MODERN_GIT for all repo-copying steps\n      t/perf: add fallback for pre-bin-wrappers versions of git\n      ewah: fix eword_t/uint64_t confusion\n      receive-pack: fix deadlock when we cannot create tmpdir\n      send-pack: extract parsing of \"unpack\" response\n      send-pack: use skip_prefix for parsing unpack status\n      send-pack: improve unpack-status error messages\n      send-pack: read \"unpack\" status even on pack-objects failure\n      send-pack: report signal death of pack-objects\n      ref-filter: move ref_cbdata definition into ref-filter.c\n      ref-filter: use contains_result enum consistently\n      ref-filter: die on parse_commit errors\n      ref-filter: use separate cache for contains_tag_algo\n      t: add an interoperability test harness\n      t/interop: add test of old clients against modern git-daemon\n      http-walker: fix buffer underflow processing remote alternates\n      add--interactive: do not expand pathspecs with ls-files\n      cherry-pick: detect bogus arguments to --mainline\n      rev-parse: use skip_prefix when parsing options\n      rev-parse: add helper for parsing \"--foo/--foo=\"\n      rev-parse: simplify parsing of ref options\n      move odb_* declarations out of git-compat-util.h\n      sha1_file.c: make pack-name helper globally accessible\n      odb_pack_keep(): stop generating keepfile name\n      replace snprintf with odb_pack_name()\n      index-pack: make pointer-alias fallbacks safer\n      sha1dc: add collision-detecting sha1 implementation\n      sha1dc: adjust header includes for git\n      sha1dc: disable safe_hash feature\n      sequencer: drop \"warning:\" when stopping for edit\n      Makefile: add DC_SHA1 knob\n      t0013: add a basic sha1 collision detection test\n      run-command: fix segfault when cleaning forked async process\n      hash-object: fix buffer reuse with --path in a subdirectory\n      prefix_filename: move docstring to header file\n      prefix_filename: drop length parameter\n      prefix_filename: return newly allocated string\n      prefix_filename: simplify windows #ifdef\n      bundle: use prefix_filename with bundle path\n      pager_in_use: use git_env_bool()\n      fast-import: use xsnprintf for writing sha1s\n      fast-import: use xsnprintf for formatting headers\n      encode_in_pack_object_header: respect output buffer length\n      pack.h: define largest possible encoded object size\n      do not check odb_mkstemp return value for errors\n      odb_mkstemp: write filename into strbuf\n      odb_mkstemp: use git_path_buf\n      Makefile: detect errors in running spatch\n      difftool: avoid strcpy\n      diff: avoid fixed-size buffer for patch-ids\n      tag: use strbuf to format tag header\n      fetch: use heap buffer to format reflog\n      avoid using fixed PATH_MAX buffers for refs\n      avoid using mksnpath for refs\n      create_branch: move msg setup closer to point of use\n      create_branch: use xstrfmt for reflog message\n      name-rev: replace static buffer with strbuf\n      receive-pack: print --pack-header directly into argv array\n      replace unchecked snprintf calls with heap buffers\n      combine-diff: replace malloc/snprintf with xstrfmt\n      convert unchecked snprintf into xsnprintf\n      transport-helper: replace checked snprintf with xsnprintf\n      gc: replace local buffer with git_path\n      daemon: use an argv_array to exec children\n      sha1_loose_object_info: return error for corrupted objects\n      index-pack: detect local corruption in collision check\n      connect.c: handle errors from split_cmdline\n      receive-pack: drop tmp_objdir_env from run_update_hook\n      receive-pack: document user-visible quarantine effects\n      refs: reject ref updates while GIT_QUARANTINE_PATH is set\n      bisect: add git_path_bisect_terms helper\n      branch: add edit_description() helper\n      use git_path_* helper functions\n      replace xstrdup(git_path(...)) with git_pathdup(...)\n      replace strbuf_addstr(git_path()) with git_path_buf()\n      am: drop \"dir\" parameter from am_state_init\n      connect.c: fix leak in handle_ssh_variant\n      doc: use https links to avoid http redirect\n      doc: replace or.cz gitwiki link with git.wiki.kernel.org\n      gitcore-tutorial: update broken link\n      docs/archimport: quote sourcecontrol.net reference\n      docs/bisect-lk2009: update nist report link\n      docs/bisect-lk2009: update java code conventions link\n      completion: optionally disable checkout DWIM\n      submodule_init: die cleanly on submodules without url defined\n      prio_queue_reverse: don't swap elements with themselves\n      shell: disallow repo names beginning with dash\n\nJiang Xin (4):\n      l10n: git.pot: v2.13.0 round 1 (96 new, 37 removed)\n      l10n: zh_CN: for git v2.13.0 l10n round 1\n      l10n: git.pot: v2.13.0 round 2 (4 new, 7 removed)\n      l10n: zh_CN: for git v2.13.0 l10n round 2\n\nJohan Hovold (1):\n      send-email: only allow one address per body tag\n\nJohannes Schindelin (23):\n      git_connect(): factor out SSH variant handling\n      rebase--helper: add a builtin helper for interactive rebases\n      rebase -i: use the rebase--helper builtin\n      rev-parse: fix several options when running in a subdirectory\n      t7006: replace dubious test\n      Travis: also test on 32-bit Linux\n      setup_git_directory(): use is_dir_sep() helper\n      t1501: demonstrate NULL pointer access with invalid GIT_WORK_TREE\n      real_pathdup(): fix callsites that wanted it to die on error\n      setup: prepare setup_discovered_git_dir() for the root directory\n      setup_git_directory_1(): avoid changing global state\n      setup: introduce the discover_git_directory() function\n      setup: make read_early_config() reusable\n      read_early_config(): avoid .git/config hack when unneeded\n      read_early_config(): really discover .git/\n      t1309: test read_early_config()\n      setup_git_directory_gently_1(): avoid die()ing\n      t1309: document cases where we would want early config not to die()\n      setup.c: mention unresolved problems\n      t7504: document regression: reword no longer calls commit-msg\n      sequencer: make commit options more extensible\n      sequencer: allow the commit-msg hooks to run during a `reword`\n      difftool: fix use-after-free\n\nJonathan Nieder (2):\n      remote helpers: avoid blind fall-back to \".git\" when setting GIT_DIR\n      has_sha1_file: don't bother if we are not in a repository\n\nJonathan Tan (5):\n      grep: do not unnecessarily query repo for \"--\"\n      upload-pack: report \"not our ref\" to client\n      http: attempt updating base URL only if no error\n      t/perf: export variable used in other blocks\n      fetch-pack: show clearer error message upon ERR\n\nJordi Mas (1):\n      l10n: Update Catalan translation\n\nJunio C Hamano (66):\n      merge: drop 'git merge <message> HEAD <commit>' syntax\n      connect: rename tortoiseplink and putty variables\n      commit.c: use strchrnul() to scan for one line\n      attr.c: use strchrnul() to scan for one line\n      attr.c: update a stale comment on \"struct match_attr\"\n      attr.c: explain the lack of attr-name syntax check in parse_attr()\n      attr.c: complete a sentence in a comment\n      attr.c: mark where #if DEBUG ends more clearly\n      attr.c: simplify macroexpand_one()\n      attr.c: tighten constness around \"git_attr\" structure\n      attr.c: plug small leak in parse_attr_line()\n      attr.c: add push_stack() helper\n      attr.c: outline the future plans by heavily commenting\n      attr: rename function and struct related to checking attributes\n      attr: (re)introduce git_check_attr() and struct attr_check\n      attr: convert git_all_attrs() to use \"struct attr_check\"\n      attr: convert git_check_attrs() callers to use the new API\n      attr: retire git_check_attrs() API\n      attr: change validity check for attribute names to use positive logic\n      ref-filter: resurrect \"strip\" as a synonym to \"lstrip\"\n      connect.c: stop conflating ssh command names and overrides\n      config: move a few helper functions up\n      config: use git_config_parse_key() in git_config_parse_parameter()\n      First batch after 2.12\n      diff: do not short-cut CHECK_SIZE_ONLY check in diff_populate_filespec()\n      branch: honor --abbrev/--no-abbrev in --list mode\n      Second batch after 2.12\n      Third batch after 2.12\n      Fourth batch after 2.12\n      Preparing for 2.12.1\n      Makefile: make DC_SHA1 the default\n      Fifth batch for 2.13\n      Git 2.12.1\n      Prepare for 2.12.2\n      Sixth batch for 2.13\n      t5615: fix a here-doc syntax error\n      t7406: correct test case for submodule-update initial population\n      tests: lint for run-away here-doc\n      branch doc: update description for `--list`\n      Git 2.12.2\n      Seventh batch for 2.13\n      sha1dc: avoid CPP macro collisions\n      Eighth batch for 2.13\n      Prepare for 2.12.3\n      Ninth batch for 2.13\n      Tenth batch for 2.13\n      remote.[ch]: parse_push_cas_option() can be static\n      contrib/git-resurrect.sh: do not write \\t for HT in sed scripts\n      Eleventh batch for 2.13\n      Twelfth batch for 2.13\n      replace: plug a memory leak\n      Git 2.13-rc0\n      Getting ready for -rc1\n      Git 2.13-rc1\n      Merging a handful of topics before -rc2\n      Git 2.13-rc2\n      Git 2.4.12\n      Git 2.5.6\n      Git 2.6.7\n      Git 2.7.5\n      Git 2.8.5\n      Git 2.9.4\n      Git 2.10.3\n      Git 2.11.2\n      Git 2.12.3\n      Git 2.13\n\nKarthik Nayak (20):\n      ref-filter: implement %(if), %(then), and %(else) atoms\n      ref-filter: include reference to 'used_atom' within 'atom_value'\n      ref-filter: implement %(if:equals=<string>) and %(if:notequals=<string>)\n      ref-filter: modify \"%(objectname:short)\" to take length\n      ref-filter: move get_head_description() from branch.c\n      ref-filter: introduce format_ref_array_item()\n      ref-filter: make %(upstream:track) prints \"[gone]\" for invalid upstreams\n      ref-filter: add support for %(upstream:track,nobracket)\n      ref-filter: make \"%(symref)\" atom work with the ':short' modifier\n      ref-filter: introduce refname_atom_parser_internal()\n      ref-filter: introduce refname_atom_parser()\n      ref-filter: make remote_ref_atom_parser() use refname_atom_parser_internal()\n      ref-filter: rename the 'strip' option to 'lstrip'\n      ref-filter: Do not abruptly die when using the 'lstrip=<N>' option\n      ref-filter: modify the 'lstrip=<N>' option to work with negative '<N>'\n      ref-filter: add an 'rstrip=<N>' option to atoms which deal with refnames\n      ref-filter: allow porcelain to translate messages in the output\n      branch, tag: use porcelain output\n      branch: use ref-filter printing APIs\n      branch: implement '--format' option\n\nKevin Willford (2):\n      name-hash: fix buffer overrun\n      t3008: skip lazy-init test on a single-core box\n\nKyle Meyer (12):\n      delete_ref: accept a reflog message argument\n      update-ref: pass reflog message to delete_ref()\n      rename_ref: replace empty message in HEAD's log\n      branch: record creation of renamed branch in HEAD's log\n      t1400: rename test descriptions to be unique\n      t1400: set core.logAllRefUpdates in \"logged by touch\" tests\n      t1400: use test_path_is_* helpers\n      t1400: remove a set of unused output files\n      t1400: use test_when_finished for cleanup\n      doc/config: grammar fixes for core.{editor,commentChar}\n      doc/revisions: remove brackets from rev^-n shorthand\n      t1400: use consistent style for test_expect_success calls\n\nLars Schneider (4):\n      travis-ci: build and test Git on Windows\n      travis-ci: add static analysis build job to run coccicheck\n      travis-ci: printf $STATUS as string\n      travis-ci: set DEVELOPER knob for Linux32 build\n\nLinus Torvalds (3):\n      pathspec magic: add '^' as alias for '!'\n      pathspec: don't error out on all-exclusionary pathspec patterns\n      mailinfo: fix in-body header continuations\n\nLuke Diamand (3):\n      git-p4: add failing test for name-rev rather than symbolic-ref\n      git-p4: add read_pipe_text() internal function\n      git-p4: don't use name-rev to get current branch\n\nMatt McCutchen (4):\n      merge-recursive: make \"CONFLICT (rename/delete)\" message show both paths\n      fetch-pack: move code to report unmatched refs to a function\n      fetch_refs_via_pack: call report_unmatched_refs\n      fetch-pack: add specific error for fetching an unadvertised object\n\nMaxim Moseychuk (2):\n      stop_progress_msg: convert xsnprintf to xstrfmt\n      bisect_next_all: convert xsnprintf to xstrfmt\n\nMichael Haggerty (52):\n      files_rename_ref(): tidy up whitespace\n      refname_is_safe(): correct docstring\n      t5505: use \"for-each-ref\" to test for the non-existence of references\n      safe_create_leading_directories_const(): preserve errno\n      safe_create_leading_directories(): set errno on SCLD_EXISTS\n      raceproof_create_file(): new function\n      lock_ref_sha1_basic(): inline constant\n      lock_ref_sha1_basic(): use raceproof_create_file()\n      rename_tmp_log(): use raceproof_create_file()\n      rename_tmp_log(): improve error reporting\n      log_ref_write(): inline function\n      log_ref_setup(): separate code for create vs non-create\n      log_ref_setup(): improve robustness against races\n      log_ref_setup(): pass the open file descriptor back to the caller\n      log_ref_write_1(): don't depend on logfile argument\n      log_ref_setup(): manage the name of the reflog file internally\n      log_ref_write_1(): inline function\n      delete_ref_loose(): derive loose reference path from lock\n      delete_ref_loose(): inline function\n      try_remove_empty_parents(): rename parameter \"name\" -> \"refname\"\n      try_remove_empty_parents(): don't trash argument contents\n      try_remove_empty_parents(): teach to remove parents of reflogs, too\n      files_transaction_commit(): clean up empty directories\n      refs: reorder some function definitions\n      refs: make some ref_store lookup functions private\n      refs: remove some unnecessary handling of submodule == \"\"\n      register_ref_store(): new function\n      refs: store submodule ref stores in a hashmap\n      refs: push the submodule attribute down\n      base_ref_store_init(): remove submodule argument\n      files_ref_store::submodule: use NULL for the main repository\n      read_loose_refs(): read refs using resolve_ref_recursively()\n      get_ref_dir(): don't call read_loose_refs() for \"refs/bisect\"\n      refs_read_raw_ref(): new function\n      refs_ref_iterator_begin(): new function\n      refs_verify_refname_available(): implement once for all backends\n      refs_verify_refname_available(): use function in more places\n      ref-cache: rename `add_ref()` to `add_ref_entry()`\n      ref-cache: rename `find_ref()` to `find_ref_entry()`\n      ref-cache: rename `remove_entry()` to `remove_entry_from_dir()`\n      refs: split `ref_cache` code into separate files\n      ref-cache: introduce a new type, ref_cache\n      refs: record the ref_store in ref_cache, not ref_dir\n      ref-cache: use a callback function to fill the cache\n      refs: handle \"refs/bisect/\" in `loose_fill_ref_dir()`\n      do_for_each_entry_in_dir(): eliminate `offset` argument\n      get_loose_ref_dir(): function renamed from get_loose_refs()\n      get_loose_ref_cache(): new function\n      cache_ref_iterator_begin(): make function smarter\n      commit_packed_refs(): use reference iteration\n      files_pack_refs(): use reference iteration\n      do_for_each_entry_in_dir(): delete function\n\nMichael J Gruber (6):\n      git-status: make porcelain more robust\n      git-prompt: add a describe style for any tags\n      mailmap: use Michael J Gruber's new address\n      describe: localize debug output fully\n      l10n: de: translate describe debug terms\n      l10n: de.po: lower case after semi-colon\n\nMichael Rappazzo (1):\n      rev-parse tests: add tests executed from a subdirectory\n\nMike Hommey (1):\n      notes: do not break note_tree structure in note_tree_consolidate()\n\nMostyn Bramley-Moore (1):\n      Documentation: document elements in \"ls-files -s\" output in order\n\nNguyễn Thái Ngọc Duy (40):\n      attr: support quoting pathname patterns in C style\n      clean: use warning_errno() when appropriate\n      config.txt: clarify multiple key values in include.path\n      config.txt: reflow the second include.path paragraph\n      config: add conditional include\n      refs.h: add forward declaration for structs used in this file\n      git-commit.txt: list post-rewrite in HOOKS section\n      files-backend: make files_log_ref_write() static\n      files-backend.c: delete dead code in files_ref_iterator_begin()\n      files-backend: delete dead code in files_init_db()\n      files-backend: add and use files_packed_refs_path()\n      files-backend: make sure files_rename_ref() always reach the end\n      files-backend: convert git_path() to strbuf_git_path()\n      files-backend: move \"logs/\" out of TMP_RENAMED_LOG\n      files-backend: add and use files_reflog_path()\n      files-backend: add and use files_ref_path()\n      files-backend: remove the use of git_path()\n      refs.c: introduce get_main_ref_store()\n      refs: rename lookup_ref_store() to lookup_submodule_ref_store()\n      refs.c: flatten get_ref_store() a bit\n      refs.c: kill register_ref_store(), add register_submodule_ref_store()\n      refs.c: make get_main_ref_store() public and use it\n      path.c: move some code out of strbuf_git_path_submodule()\n      refs: move submodule code out of files-backend.c\n      files-backend: replace submodule_allowed check in files_downcast()\n      refs: rename get_ref_store() to get_submodule_ref_store() and make it public\n      refs: add new ref-store api\n      refs: new transaction related ref-store api\n      files-backend: avoid ref api targeting main ref store\n      refs: delete pack_refs() in favor of refs_pack_refs()\n      t/helper: add test-ref-store to test ref-store functions\n      t1405: some basic tests on main ref store\n      t1406: new tests for submodule ref store\n      refs.h: add a note about sorting order of for_each_ref_*\n      path.c: and an option to call real_path() in expand_user_path()\n      config: resolve symlinks in conditional include's patterns\n      config: prepare to pass more info in git_config_with_options()\n      config: handle conditional include when $GIT_DIR is not set up\n      config: correct file reading order in read_early_config()\n      worktree add: add --lock option\n\nPatrick Steinhardt (8):\n      mailmap: add Patrick Steinhardt's work address\n      urlmatch: enable normalization of URLs with globs\n      urlmatch: split host and port fields in `struct url_info`\n      urlmatch: include host in urlmatch ranking\n      urlmatch: allow globbing for the URL host part\n      docs/diffcore: fix grammar in diffcore-rename header\n      docs/diffcore: unquote \"Complete Rewrites\" in headers\n      pathspec: honor `PATHSPEC_PREFIX_ORIGIN` with empty prefix\n\nPeter Krefting (3):\n      l10n: sv.po: Update Swedish translation (3199t0f0u)\n      i18n: read-cache: typofix\n      l10n: sv.po: Update Swedish translation (3195t0f0u)\n\nPrathamesh Chavan (1):\n      t2027: avoid using pipes\n\nQuentin Pradet (1):\n      git-bisect.txt: add missing word\n\nRalf Thielow (4):\n      submodule.c: add missing ' in error messages\n      git-add--interactive.perl: add missing dot in a message\n      l10n: de.po: update German translation\n      l10n: de.po: translate 4 new messages\n\nRamsay Jones (3):\n      wrapper.c: remove unused git_mkstemp() function\n      wrapper.c: remove unused gitmkstemps() function\n      name-hash: add test-lazy-init-name-hash to .gitignore\n\nRay Chen (1):\n      l10n: zh_CN: review for git v2.13.0 l10n round 1\n\nRené Genz (3):\n      doc: update SubmittingPatches\n      fix minor typos\n      doc: git-pull.txt use US spelling, fix minor typo\n\nRené Scharfe (18):\n      sha1_file: release fallback base's memory in unpack_entry()\n      cocci: use ALLOC_ARRAY\n      strbuf: add strbuf_add_real_path()\n      commit: be more precise when searching for headers\n      commit: don't check for space twice when looking for header\n      log-tree: factor out fmt_output_email_subject()\n      pretty: use fmt_output_email_subject()\n      blame: move blame_entry duplication to add_blame_entry()\n      receive-pack: simplify run_update_post_hook()\n      http-push: don't check return value of lookup_unknown_object()\n      revision: remove declaration of path_name()\n      shortlog: don't set after_subject to an empty string\n      strbuf: support long paths w/o read rights in strbuf_getcwd() on FreeBSD\n      test-read-cache: setup git dir\n      am: close stream on error, but not stdin\n      files_for_each_reflog_ent_reverse(): close stream and free strbuf on error\n      server-info: avoid calling fclose(3) twice in update_info_file()\n      use HOST_NAME_MAX to size buffers for gethostname(2)\n\nRoss Lagerwall (1):\n      remote: ignore failure to remove missing branch.<name>.merge\n\nSZEDER Gábor (44):\n      completion: improve __git_refs()'s in-code documentation\n      completion tests: don't add test cruft to the test repository\n      completion tests: make the $cur variable local to the test helper functions\n      completion tests: consolidate getting path of current working directory\n      completion tests: check __gitdir()'s output in the error cases\n      completion tests: add tests for the __git_refs() helper function\n      completion: ensure that the repository path given on the command line exists\n      completion: fix most spots not respecting 'git --git-dir=<path>'\n      completion: respect 'git --git-dir=<path>' when listing remote refs\n      completion: list refs from remote when remote's name matches a directory\n      completion: don't list 'HEAD' when trying refs completion outside of a repo\n      completion: list short refs from a remote given as a URL\n      completion: don't offer commands when 'git --opt' needs an argument\n      completion: fix completion after 'git -C <path>'\n      rev-parse: add '--absolute-git-dir' option\n      completion: respect 'git -C <path>'\n      completion: don't use __gitdir() for git commands\n      completion: consolidate silencing errors from git commands\n      completion: don't guard git executions with __gitdir()\n      completion: extract repository discovery from __gitdir()\n      completion: cache the path to the repository\n      completion: restore removed line continuating backslash\n      completion: remove redundant __gitcomp_nl() options from _git_commit()\n      tests: create an interactive gdb session with the 'debug' helper\n      tests: make the 'test_pause' helper work in non-verbose mode\n      pickaxe: fix segfault with '-S<...> --pickaxe-regex'\n      completion: wrap __git_refs() for better option parsing\n      completion: support completing full refs after '--option=refs/<TAB>'\n      completion: support completing fully qualified non-fast-forward refspecs\n      completion: support excluding full refs\n      completion: don't disambiguate tags and branches\n      completion: don't disambiguate short refs\n      completion: let 'for-each-ref' and 'ls-remote' filter matching refs\n      completion: let 'for-each-ref' strip the remote name from remote branches\n      completion: let 'for-each-ref' filter remote branches for 'checkout' DWIMery\n      completion: let 'for-each-ref' sort remote branches for 'checkout' DWIMery\n      completion: fill COMPREPLY directly when completing refs\n      completion: fill COMPREPLY directly when completing fetch refspecs\n      completion: speed up branch and tag completion\n      refs.c: use skip_prefix() in prettify_refname()\n      completion: put matching ctags symbol names directly into COMPREPLY\n      completion: extract completing ctags symbol names into helper function\n      completion: offer ctags symbol names for 'git log -S', '-G' and '-L:'\n      t6500: wait for detached auto gc at the end of the test script\n\nSantiago Torres (1):\n      t7004, t7030: fix here-doc syntax errors\n\nSebastian Schuberth (5):\n      contrib: git-remote-{bzr,hg} placeholders don't need Python\n      sha1_file: remove an used fd variable\n      submodule: remove a superfluous second check for the \"new\" variable\n      gitmodules: clarify what history depth a shallow clone has\n      gitmodules: clarify the ignore option values\n\nSegev Finer (2):\n      connect: handle putty/plink also in GIT_SSH_COMMAND\n      connect: Add the envvar GIT_SSH_VARIANT and ssh.variant config\n\nSergey Ryazanov (2):\n      http: honor empty http.proxy option to bypass proxy\n      http: fix the silent ignoring of proxy misconfiguraion\n\nSimon Ruderich (1):\n      githooks.txt: clarify push hooks are always executed in $GIT_DIR\n\nStefan Beller (36):\n      Documentation: fix a typo\n      refs: parse_hide_refs_config to use parse_config_key\n      submodule init: warn about falling back to a local path\n      rev-parse: add --show-superproject-working-tree\n      submodule-config: correct error reporting for invalid ignore value\n      connect_work_tree_and_git_dir: safely create leading directories\n      lib-submodule-update.sh: reorder create_lib_submodule_repo\n      lib-submodule-update.sh: do not use ./. as submodule remote\n      lib-submodule-update: teach test_submodule_content the -C <dir> flag\n      lib-submodule-update.sh: replace sha1 by hash\n      lib-submodule-update.sh: define tests for recursing into submodules\n      make is_submodule_populated gently\n      update submodules: add submodule config parsing\n      update submodules: add a config option to determine if submodules are updated\n      submodules: introduce check to see whether to touch a submodule\n      update submodules: move up prepare_submodule_repo_env\n      submodule.c: get_super_prefix_or_empty\n      update submodules: add submodule_move_head\n      unpack-trees: pass old oid to verify_clean_submodule\n      unpack-trees: check if we can perform the operation for submodules\n      entry.c: create submodules when interesting\n      builtin/checkout: add --recurse-submodules switch\n      builtin/read-tree: add --recurse-submodules switch\n      wt-status: simplify by using for_each_string_list_item\n      t3600: rename test to describe its functionality\n      builtin/describe: introduce --broken flag\n      submodule.c: use argv_array in is_submodule_modified\n      submodule.c: factor out early loop termination in is_submodule_modified\n      submodule.c: convert is_submodule_modified to use strbuf_getwholeline\n      submodule.c: port is_submodule_modified to use porcelain 2\n      submodule.c: stricter checking for submodules in is_submodule_modified\n      short status: improve reporting for submodule changes\n      unpack-trees.c: align submodule error message to the other error messages\n      submodule.c: correctly handle nested submodules in is_submodule_modified\n      diff: submodule inline diff to initialize env array.\n      submodule--helper: fix typo in is_active error message\n\nStephen Hicks (1):\n      rebase -i: reread the todo list if `exec` touched it\n\nThomas Gummerer (9):\n      stash: introduce push verb\n      stash: add test for the create command line arguments\n      stash: refactor stash_create\n      stash: teach 'push' (and 'create_stash') to honor pathspec\n      stash: use stash_push for no verb form\n      stash: allow pathspecs in the no verb form\n      stash: don't show internal implementation details\n      stash: pass the pathspec argument to git reset\n      stash: keep untracked files intact in stash -k\n\nTorsten Bögershausen (1):\n      gitattributes.txt: document how to normalize the line endings\n\nTrần Ngọc Quân (2):\n      l10n: vi.po(3198t): Updated Vietnamese translation for v2.13.0-rc0\n      l10n: vi.po(3195t): Update translation for v2.13.0 round 2\n\nValery Tolstov (1):\n      submodule--helper.c: remove duplicate code\n\nVasco Almeida (1):\n      l10n: pt_PT: update Portuguese translation\n\nVegard Nossum (2):\n      xdiff -W: relax end-of-file function detection\n      line-log: use COPY_ARRAY to fix mis-sized memcpy\n\nbrian m. carlson (41):\n      hex: introduce parse_oid_hex\n      builtin/commit: convert to struct object_id\n      builtin/diff-tree: convert to struct object_id\n      builtin/describe: convert to struct object_id\n      builtin/fast-export: convert to struct object_id\n      builtin/fmt-merge-message: convert to struct object_id\n      builtin/grep: convert to struct object_id\n      builtin/branch: convert to struct object_id\n      builtin/clone: convert to struct object_id\n      builtin/merge: convert to struct object_id\n      Convert remaining callers of resolve_refdup to object_id\n      builtin/replace: convert to struct object_id\n      reflog-walk: convert struct reflog_info to struct object_id\n      refs: convert each_reflog_ent_fn to struct object_id\n      refs: simplify parsing of reflog entries\n      sha1_file: introduce an nth_packed_object_oid function\n      Convert object iteration callbacks to struct object_id\n      builtin/merge-base: convert to struct object_id\n      wt-status: convert to struct object_id\n      hash.h: move SHA-1 implementation selection into a header file\n      Define new hash-size constants for allocating memory\n      Convert GIT_SHA1_HEXSZ used for allocation to GIT_MAX_HEXSZ\n      Convert GIT_SHA1_RAWSZ used for allocation to GIT_MAX_RAWSZ\n      builtin/diff: convert to struct object_id\n      builtin/pull: convert portions to struct object_id\n      builtin/receive-pack: fix incorrect pointer arithmetic\n      builtin/receive-pack: convert portions to struct object_id\n      fsck: convert init_skiplist to struct object_id\n      parse-options-cb: convert sha1_array_append caller to struct object_id\n      test-sha1-array: convert most code to struct object_id\n      sha1_name: convert struct disambiguate_state to object_id\n      sha1_name: convert disambiguate_hint_fn to take object_id\n      submodule: convert check_for_new_submodule_commits to object_id\n      builtin/pull: convert to struct object_id\n      sha1-array: convert internal storage for struct sha1_array to object_id\n      Make sha1_array_append take a struct object_id *\n      Convert remaining callers of sha1_array_lookup to object_id\n      Convert sha1_array_lookup to take struct object_id\n      Convert sha1_array_for_each_unique and for_each_abbrev to object_id\n      Rename sha1_array to oid_array\n      Documentation: update and rename api-sha1-array.txt\n\nÆvar Arnfjörð Bjarmason (33):\n      cvs tests: skip tests that call \"cvs commit\" when running as root\n      gitweb tests: change confusing \"skip_all\" phrasing\n      gitweb tests: skip tests when we don't have Time::HiRes\n      doc: change erroneous --[no]-whatever into --[no-]whatever\n      push: mention \"push.default=tracking\" in the documentation\n      tag doc: move the description of --[no-]merged earlier\n      tag doc: split up the --[no-]merged documentation\n      tag doc: reword --[no-]merged to talk about commits, not tips\n      ref-filter: make combining --merged & --no-merged an error\n      doc/SubmittingPatches: clarify the casing convention for \"area: change...\"\n      t/README: link to metacpan.org, not search.cpan.org\n      t/README: change \"Inside <X> part\" to \"Inside the <X> part\"\n      ref-filter: add test for --contains on a non-commit\n      tag: remove a TODO item from the test suite\n      tag tests: fix a typo in a test description\n      for-each-ref: partly change <object> to <commit> in help\n      branch doc: change `git branch <pattern>` to use `<branchname>`\n      tag: add more incompatibles mode tests\n      parse-options: add OPT_NONEG to the \"contains\" option\n      tag: change misleading --list <pattern> documentation\n      tag: implicitly supply --list given another list-like option\n      tag: change --point-at to default to HEAD\n      ref-filter: add --no-contains option to tag/branch/for-each-ref\n      ref-filter: reflow recently changed branch/tag/for-each-ref docs\n      tag: add tests for --with and --without\n      t/README: clarify the test_have_prereq documentation\n      doc/SubmittingPatches: show how to get a CLI commit summary\n      rev-parse: match @{upstream}, @{u} and @{push} case-insensitively\n      perl: regenerate perl.mak if perl -V changes\n      grep: plug a trivial memory leak\n      push: document & test --force-with-lease with multiple remotes\n      completion: expand \"push --delete <remote> <ref>\" for refs on that <remote>\n      t/perf: correctly align non-ASCII descriptions in output\n\n"}]}