1Git 2.7 Release Notes 2===================== 3 4Updates since v2.6 5------------------ 6 7UI, Workflows & Features 8 9 * "git remote" learned "get-url" subcommand to show the URL for a 10 given remote name used for fetching and pushing. 11 12 * There was no way to defeat a configured rebase.autostash variable 13 from the command line, as "git rebase --no-autostash" was missing. 14 15 * "git log --date=local" used to only show the normal (default) 16 format in the local timezone. The command learned to take 'local' 17 as an instruction to use the local timezone with other formats, 18 19 * The refs used during a "git bisect" session is now per-worktree so 20 that independent bisect sessions can be done in different worktrees 21 created with "git worktree add". 22 23 * Users who are too busy to type three extra keystrokes to ask for 24 "git stash show -p" can now set stash.showPatch configuration 25 varible to true to always see the actual patch, not just the list 26 of paths affected with feel for the extent of damage via diffstat. 27 28 * "quiltimport" allows to specify the series file by honoring the 29 $QUILT_SERIES environment and also --series command line option. 30 31 * The use of 'good/bad' in "git bisect" made it confusing to use when 32 hunting for a state change that is not a regression (e.g. bugfix). 33 The command learned 'old/new' and then allows the end user to 34 say e.g. "bisect start --term-old=fast --term-new=slow" to find a 35 performance regression. 36 37 * "git interpret-trailers" can now run outside of a Git repository. 38 39 * "git p4" learned to reencode the pathname it uses to communicate 40 with the p4 depot with a new option. 41 42 * Give progress meter to "git filter-branch". 43 44 * Allow a later "!/abc/def" to override an earlier "/abc" that 45 appears in the same .gitignore file to make it easier to express 46 "everything in /abc directory is ignored, except for ...". 47 48 * Teach "git p4" to send large blobs outside the repository by 49 talking to Git LFS. 50 51 * Prepare for Git on-disk repository representation to undergo 52 backward incompatible changes by introducing a new repository 53 format version "1", with an extension mechanism. 54 55 * "git worktree" learned a "list" subcommand. 56 57 * "git clone --dissociate" learned that it can be used even when 58 "--reference" was not used at the same time. 59 60 * "git blame" learnt to take "--first-parent" and "--reverse" at the 61 same time when it makes sense. 62 63 * "git checkout" did not follow the usual "--[no-]progress" 64 convention and implemented only "--quiet" that is essentially 65 a superset of "--no-progress". Extend the command to support the 66 usual "--[no-]progress". 67 68 * The semantics of tranfer.hideRefs configuration variable have been 69 extended to work better with the ref "namespace" feature that lets 70 you throw unrelated bunches of repositories in a single physical 71 repository and virtually serve them as separate ones. 72 73 * send-email config variables whose values are pathnames now go 74 through the ~username/ expansion. 75 76 * bash completion learnt to TAB-complete recipient addresses given 77 to send-email. 78 79 * The credential-cache daemon can be told to ignore SIGHUP to work 80 around issue when running Git from inside emacs. 81 82 83Performance, Internal Implementation, Development Support etc. 84 85 * The infrastructure to rewrite "git submodule" in C is being built 86 incrementally. Let's polish these early parts well enough and make 87 them graduate to 'next' and 'master', so that the more involved 88 follow-up can start cooking on a solid ground. 89 90 * Some features from "git tag -l" and "git branch -l" have been made 91 available to "git for-each-ref" so that eventually the unified 92 implementation can be shared across all three. The version merged 93 to the 'master' branch earlier had a performance regression in "tag 94 --contains", which has since been corrected. 95 96 * Because "test_when_finished" in our test framework queues the 97 clean-up tasks to be done in a shell variable, it should not be 98 used inside a subshell. Add a mechanism to allow 'bash' to catch 99 such uses, and fix the ones that were found. 100 101 * The debugging infrastructure for pkt-line based communication has 102 been improved to mark the side-band communication specifically. 103 (merge fd89433 jk/async-pkt-line later to maint). 104 105 * Update "git branch" that list existing branches, using the 106 ref-filter API that is shared with "git tag" and "git 107 for-each-ref". 108 109 * The test for various line-ending conversions has been enhanced. 110 111 * A few test scripts around "git p4" have been improved for 112 portability. 113 114 * Many allocations that is manually counted (correctly) that are 115 followed by strcpy/sprintf have been replaced with a less error 116 prone constructs such as xstrfmt. 117 118 * The internal stripspace() function has been moved to where it 119 logically belongs to, i.e. strbuf API, and the command line parser 120 of "git stripspace" has been updated to use the parse_options API. 121 122 * "git am" used to spawn "git mailinfo" via run_command() API once 123 per each patch, but learned to make a direct call to mailinfo() 124 instead. 125 126 * The implementation of "git mailinfo" was refactored so that a 127 mailinfo() function can be directly called from inside a process. 128 129 * With a "debug" helper, debugging of a single "git" invocation in 130 our test scripts has become a lot easier. 131 132 * The "configure" script did not test for -lpthread correctly, which 133 upset some linkers. 134 135 * Cross completed task off of subtree project's todo list. 136 137 * Test cleanups for the subtree project. 138 139 * Clean up style in an ancient test t9300. 140 141 * Work around some test flakiness with p4d. 142 143 * Fsck did not correctly detect a NUL-truncated header in a tag. 144 145 * Use a safer behavior when we hit errors verifying remote certificates. 146 147 * Speed up filter-branch for cases where we only care about rewriting 148 commits, not tree data. 149 150 * The parse-options API has been updated to make "-h" command line 151 option work more consistently in all commands. 152 153 * "git svn rebase/mkdirs" got optimized by keeping track of empty 154 directories better. 155 156 * Fix some racy client/server tests by treating SIGPIPE the same as a 157 normal non-zero exit. 158 159 * The necessary infrastructure to build topics using the free Travis 160 CI has been added. Developers forking from this topic (and enabling 161 Travis) can do their own builds, and we can turn on auto-builds for 162 git/git (including build-status for pull requests that people 163 open). 164 165 166Also contains various documentation updates and code clean-ups. 167 168 169Fixes since v2.6 170---------------- 171 172Unless otherwise noted, all the fixes since v2.6 in the maintenance 173track are contained in this release (see the maintenance releases' 174notes for details). 175 176 * Very small number of options take a parameter that is optional 177 (which is not a great UI element as they can only appear at the end 178 of the command line). Add notice to documentation of each and 179 every one of them. 180 181 * "git blame --first-parent v1.0..v2.0" was not rejected but did not 182 limit the blame to commits on the first parent chain. 183 184 * "git subtree" (in contrib/) now can take whitespaces in the 185 pathnames, not only in the in-tree pathname but the name of the 186 directory that the repository is in. 187 188 * The ssh transport, just like any other transport over the network, 189 did not clear GIT_* environment variables, but it is possible to 190 use SendEnv and AcceptEnv to leak them to the remote invocation of 191 Git, which is not a good idea at all. Explicitly clear them just 192 like we do for the local transport. 193 194 * Correct "git p4 --detect-labels" so that it does not fail to create 195 a tag that points at a commit that is also being imported. 196 197 * The Makefile always runs the library archiver with hardcoded "crs" 198 options, which was inconvenient for exotic platforms on which 199 people want to use programs with totally different set of command 200 line options. 201 202 * Customization to change the behaviour with "make -w" and "make -s" 203 in our Makefile was broken when they were used together. 204 205 * Allocation related functions and stdio are unsafe things to call 206 inside a signal handler, and indeed killing the pager can cause 207 glibc to deadlock waiting on allocation mutex as our signal handler 208 tries to free() some data structures in wait_for_pager(). Reduce 209 these unsafe calls. 210 211 * The way how --ref/--notes to specify the notes tree reference are 212 DWIMmed was not clearly documented. 213 214 * "git gc" used to barf when a symbolic ref has gone dangling 215 (e.g. the branch that used to be your upstream's default when you 216 cloned from it is now gone, and you did "fetch --prune"). 217 218 * "git clone --dissociate" runs a big "git repack" process at the 219 end, and it helps to close file descriptors that are open on the 220 packs and their idx files before doing so on filesystems that 221 cannot remove a file that is still open. 222 223 * Description of the "log.follow" configuration variable in "git log" 224 documentation is now also copied to "git config" documentation. 225 226 * "git rebase -i" had a minor regression recently, which stopped 227 considering a line that begins with an indented '#' in its insn 228 sheet not a comment. Further, the code was still too picky on 229 Windows where CRLF left by the editor is turned into a trailing CR 230 on the line read via the "read" built-in command of bash. Both of 231 these issues are now fixed. 232 233 * After "git checkout --detach", "git status" reported a fairly 234 useless "HEAD detached at HEAD", instead of saying at which exact 235 commit. 236 237 * When "git send-email" wanted to talk over Net::SMTP::SSL, 238 Net::Cmd::datasend() did not like to be fed too many bytes at the 239 same time and failed to send messages. Send the payload one line 240 at a time to work around the problem. 241 242 * When "git am" was rewritten as a built-in, it stopped paying 243 attention to user.signingkey, which was fixed. 244 245 * It was not possible to use a repository-lookalike created by "git 246 worktree add" as a local source of "git clone". 247 248 * On a case insensitive filesystems, setting GIT_WORK_TREE variable 249 using a random cases that does not agree with what the filesystem 250 thinks confused Git that it wasn't inside the working tree. 251 252 * Performance-measurement tests did not work without an installed Git. 253 254 * A test script for the HTTP service had a timing dependent bug, 255 which was fixed. 256 257 * There were some classes of errors that "git fsck" diagnosed to its 258 standard error that did not cause it to exit with non-zero status. 259 260 * Work around "git p4" failing when the P4 depot records the contents 261 in UTF-16 without UTF-16 BOM. 262 263 * When "git gc --auto" is backgrounded, its diagnosis message is 264 lost. Save it to a file in $GIT_DIR and show it next time the "gc 265 --auto" is run. 266 267 * The submodule code has been taught to work better with separate 268 work trees created via "git worktree add". 269 270 * "git gc" is safe to run anytime only because it has the built-in 271 grace period to protect young objects. In order to run with no 272 grace period, the user must make sure that the repository is 273 quiescent. 274 275 * A recent "filter-branch --msg-filter" broke skipping of the commit 276 object header, which is fixed. 277 278 * The normalize_ceiling_entry() function does not muck with the end 279 of the path it accepts, and the real world callers do rely on that, 280 but a test insisted that the function drops a trailing slash. 281 282 * A test for interaction between untracked cache and sparse checkout 283 added in Git 2.5 days were flaky. 284 285 * A couple of commands still showed "[options]" in their usage string 286 to note where options should come on their command line, but we 287 spell that "[<options>]" in most places these days. 288 289 * The synopsis text and the usage string of subcommands that read 290 list of things from the standard input are often shown as if they 291 only take input from a file on a filesystem, which was misleading. 292 293 * "git am -3" had a small regression where it is aborted in its error 294 handling codepath when underlying merge-recursive failed in certain 295 ways, as it assumed that the internal call to merge-recursive will 296 never die, which is not the case (yet). 297 298 * The linkage order of libraries was wrong in places around libcurl. 299 300 * The name-hash subsystem that is used to cope with case insensitive 301 filesystems keeps track of directories and their on-filesystem 302 cases for all the paths in the index by holding a pointer to a 303 randomly chosen cache entry that is inside the directory (for its 304 ce->ce_name component). This pointer was not updated even when the 305 cache entry was removed from the index, leading to use after free. 306 This was fixed by recording the path for each directory instead of 307 borrowing cache entries and restructuring the API somewhat. 308 309 * "git merge-file" tried to signal how many conflicts it found, which 310 obviously would not work well when there are too many of them. 311 312 * The error message from "git blame --contents --reverse" incorrectly 313 talked about "--contents --children". 314 315 * "git imap-send" did not compile well with older version of cURL library. 316 317 * Merging a branch that removes a path and another that changes the 318 mode bits on the same path should have conflicted at the path, but 319 it didn't and silently favoured the removal. 320 321 * "git --literal-pathspecs add -u/-A" without any command line 322 argument misbehaved ever since Git 2.0. 323 324 * "git daemon" uses "run_command()" without "finish_command()", so it 325 needs to release resources itself, which it forgot to do. 326 327 * "git status --branch --short" accessed beyond the constant string 328 "HEAD", which has been corrected. 329 330 * We peek objects from submodule's object store by linking it to the 331 list of alternate object databases, but the code to do so forgot to 332 correctly initialize the list. 333 334 * The code to prepare the working tree side of temporary directory 335 for the "dir-diff" feature forgot that symbolic links need not be 336 copied (or symlinked) to the temporary area, as the code already 337 special cases and overwrites them. Besides, it was wrong to try 338 computing the object name of the target of symbolic link, which may 339 not even exist or may be a directory. 340 341 * A Range: request can be responded with a full response and when 342 asked properly libcurl knows how to strip the result down to the 343 requested range. However, we were hand-crafting a range request 344 and it did not kick in. 345 346 * Having a leftover .idx file without corresponding .pack file in 347 the repository hurts performance; "git gc" learned to prune them. 348 349 * Apple's common crypto implementation of SHA1_Update() does not take 350 more than 4GB at a time, and we now have a compile-time workaround 351 for it. 352 353 * Produce correct "dirty" marker for shell prompts, even when we 354 are on an orphan or an unborn branch. 355 356 * A build without NO_IPv6 used to use gethostbyname() when guessing 357 user's hostname, instead of getaddrinfo() that is used in other 358 codepaths in such a build. 359 (merge 00bce77 ep/ident-with-getaddrinfo later to maint). 360 361 * The exit code of git-fsck didnot reflect some types of errors found 362 in packed objects, which has been corrected. 363 (merge 8c24d83 dt/fsck-verify-pack-error later to maint). 364 365 * Code clean-up, minor fixes etc. 366 (merge 15ed07d jc/rerere later to maint). 367 (merge 147875f sb/submodule-config-parse later to maint).