1Git 2.14 Release Notes 2====================== 3 4Backward compatibility notes. 5 6 * Use of an empty string as a pathspec element that is used for 7 'everything matches' is still warned and Git asks users to use a 8 more explicit '.' for that instead. The hope is that existing 9 users will not mind this change, and eventually the warning can be 10 turned into a hard error, upgrading the deprecation into removal of 11 this (mis)feature. That is not scheduled to happen in the upcoming 12 release (yet). 13 14 * Git now avoids blindly falling back to ".git" when the setup 15 sequence said we are _not_ in Git repository. A corner case that 16 happens to work right now may be broken by a call to die("BUG"). 17 We've tried hard to locate such cases and fixed them, but there 18 might still be cases that need to be addressed--bug reports are 19 greatly appreciated. 20 21 * The experiment to improve the hunk-boundary selection of textual 22 diff output has finished, and the "indent heuristics" has now 23 become the default. 24 25 26Updates since v2.13 27------------------- 28 29UI, Workflows & Features 30 31 * The colors in which "git status --short --branch" showed the names 32 of the current branch and its remote-tracking branch are now 33 configurable. 34 35 * "git clone" learned the "--no-tags" option not to fetch all tags 36 initially, and also set up the tagopt not to follow any tags in 37 subsequent fetches. 38 39 * "git archive --format=zip" learned to use zip64 extension when 40 necessary to go beyond the 4GB limit. 41 42 * "git reset" learned "--recurse-submodules" option. 43 44 * "git diff --submodule=diff" now recurses into nested submodules. 45 46 * "git repack" learned to accept the --threads=<n> option and pass it 47 to pack-objects. 48 49 * "git send-email" learned to run sendemail-validate hook to inspect 50 and reject a message before sending it out. 51 52 * There is no good reason why "git fetch $there $sha1" should fail 53 when the $sha1 names an object at the tip of an advertised ref, 54 even when the other side hasn't enabled allowTipSHA1InWant. 55 56 * The recently introduced "[includeIf "gitdir:$dir"] path=..." 57 mechanism has further been taught to take symlinks into account. 58 The directory "$dir" specified in "gitdir:$dir" may be a symlink to 59 a real location, not something that $(getcwd) may return. In such 60 a case, a realpath of "$dir" is compared with the real path of the 61 current repository to determine if the contents from the named path 62 should be included. 63 64 * Make the "indent" heuristics the default in "diff" and diff.indentHeuristics 65 configuration variable an escape hatch for those who do no want it. 66 67 * Many commands learned to pay attention to submodule.recurse 68 configuration. 69 70 * The convention for a command line is to follow "git cmdname 71 --options" with revisions followed by an optional "--" 72 disambiguator and then finally pathspecs. When "--" is not there, 73 we make sure early ones are all interpretable as revs (and do not 74 look like paths) and later ones are the other way around. A 75 pathspec with "magic" (e.g. ":/p/a/t/h" that matches p/a/t/h from 76 the top-level of the working tree, no matter what subdirectory you 77 are working from) are conservatively judged as "not a path", which 78 required disambiguation more often. The command line parser 79 learned to say "it's a pathspec" a bit more often when the syntax 80 looks like so. 81 82 * Update "perl-compatible regular expression" support to enable JIT 83 and also allow linking with the newer PCRE v2 library. 84 85 * "filter-branch" learned a pseudo filter "--setup" that can be used 86 to define common functions/variables that can be used by other 87 filters. 88 89 * Using "git add d/i/r" when d/i/r is the top of the working tree of 90 a separate repository would create a gitlink in the index, which 91 would appear as a not-quite-initialized submodule to others. We 92 learned to give warnings when this happens. 93 94 * "git status" learned to optionally give how many stash entries the 95 user has in its output. 96 97 * "git status" has long shown essentially the same message as "git 98 commit"; the message it gives while preparing for the root commit, 99 i.e. "Initial commit", was hard to understand for some new users. 100 Now it says "No commits yet" to stress more on the current status 101 (rather than the commit the user is preparing for, which is more in 102 line with the focus of "git commit"). 103 104Performance, Internal Implementation, Development Support etc. 105 106 * The default packed-git limit value has been raised on larger 107 platforms to save "git fetch" from a (recoverable) failure while 108 "gc" is running in parallel. 109 110 * Code to update the cache-tree has been tightened so that we won't 111 accidentally write out any 0{40} entry in the tree object. 112 113 * Attempt to allow us notice "fishy" situation where we fail to 114 remove the temporary directory used during the test. 115 116 * Travis CI gained a task to format the documentation with both 117 AsciiDoc and AsciiDoctor. 118 119 * Some platforms have ulong that is smaller than time_t, and our 120 historical use of ulong for timestamp would mean they cannot 121 represent some timestamp that the platform allows. Invent a 122 separate and dedicated timestamp_t (so that we can distingiuish 123 timestamps and a vanilla ulongs, which along is already a good 124 move), and then declare uintmax_t is the type to be used as the 125 timestamp_t. 126 127 * We can trigger Windows auto-build tester (credits: Dscho & 128 Microsoft) from our existing Travis CI tester now. 129 130 * Conversion from uchar[20] to struct object_id continues. 131 132 * Simplify parse_pathspec() codepath and stop it from looking at the 133 default in-core index. 134 135 * Add perf-test for wildmatch. 136 137 * Code from "conversion using external process" codepath has been 138 extracted to a separate sub-process.[ch] module. 139 140 * When "git checkout", "git merge", etc. manipulates the in-core 141 index, various pieces of information in the index extensions are 142 discarded from the original state, as it is usually not the case 143 that they are kept up-to-date and in-sync with the operation on the 144 main index. The untracked cache extension is copied across these 145 operations now, which would speed up "git status" (as long as the 146 cache is properly invalidated). 147 148 * The internal implementation of "git grep" has seen some clean-up. 149 150 * Update the C style recommendation for notes for translators, as 151 recent versions of gettext tools can work with our style of 152 multi-line comments. 153 154 * The implementation of "ref" API around the "packed refs" have been 155 cleaned up, in preparation for further changes. 156 157 * The internal logic used in "git blame" has been libified to make it 158 easier to use by cgit. 159 160 * Our code often opens a path to an optional file, to work on its 161 contents when we can successfully open it. We can ignore a failure 162 to open if such an optional file does not exist, but we do want to 163 report a failure in opening for other reasons (e.g. we got an I/O 164 error, or the file is there, but we lack the permission to open). 165 166 The exact errors we need to ignore are ENOENT (obviously) and 167 ENOTDIR (less obvious). Instead of repeating comparison of errno 168 with these two constants, introduce a helper function to do so. 169 170 * We often try to open a file for reading whose existence is 171 optional, and silently ignore errors from open/fopen; report such 172 errors if they are not due to missing files. 173 174 * When an existing repository is used for t/perf testing, we first 175 create bit-for-bit copy of it, which may grab a transient state of 176 the repository and freeze it into the repository used for testing, 177 which then may cause Git operations to fail. Single out "the index 178 being locked" case and forcibly drop the lock from the copy. 179 180 * Three instances of the same helper function have been consolidated 181 to one. 182 183 * "fast-import" uses a default pack chain depth that is consistent 184 with other parts of the system. 185 186 * A new test to show the interaction between the pattern [^a-z] 187 (which matches '/') and a slash in a path has been added. The 188 pattern should not match the slash with "pathmatch", but should 189 with "wildmatch". 190 191 * The 'diff-highlight' program (in contrib/) has been restructured 192 for easier reuse by an external project 'diff-so-fancy'. 193 (merge 0c977dbc81 jk/diff-highlight-module later to maint). 194 195 * A common pattern to free a piece of memory and assign NULL to the 196 pointer that used to point at it has been replaced with a new 197 FREE_AND_NULL() macro. 198 199 * Traditionally, the default die() routine had a code to prevent it 200 from getting called multiple times, which interacted badly when a 201 threaded program used it (one downside is that the real error may 202 be hidden and instead the only error message given to the user may 203 end up being "die recursion detected", which is not very useful). 204 205 206Also contains various documentation updates and code clean-ups. 207 208 209Fixes since v2.13 210----------------- 211 212Unless otherwise noted, all the fixes since v2.13 in the maintenance 213track are contained in this release (see the maintenance releases' 214notes for details). 215 216 * "git gc" did not interact well with "git worktree"-managed 217 per-worktree refs. 218 219 * "git cherry-pick" and other uses of the sequencer machinery 220 mishandled a trailer block whose last line is an incomplete line. 221 This has been fixed so that an additional sign-off etc. are added 222 after completing the existing incomplete line. 223 224 * The codepath in "git am" that is used when running "git rebase" 225 leaked memory held for the log message of the commits being rebased. 226 227 * "git clone --config var=val" is a way to populate the 228 per-repository configuration file of the new repository, but it did 229 not work well when val is an empty string. This has been fixed. 230 231 * Setting "log.decorate=false" in the configuration file did not take 232 effect in v2.13, which has been corrected. 233 234 * A few codepaths in "checkout" and "am" working on an unborn branch 235 tried to access an uninitialized piece of memory. 236 237 * The Web interface to gmane news archive is long gone, even though 238 the articles are still accessible via NTTP. Replace the links with 239 ones to public-inbox.org. Because their message identification is 240 based on the actual message-id, it is likely that it will be easier 241 to migrate away from it if/when necessary. 242 243 * The receive-pack program now makes sure that the push certificate 244 records the same set of push options used for pushing. 245 246 * Tests have been updated to pass under GETTEXT_POISON (a mechanism 247 to ensure that output strings that should not be translated are 248 not translated by mistake), and TravisCI is told to run them. 249 250 * "git checkout --recurse-submodules" did not quite work with a 251 submodule that itself has submodules. 252 253 * "pack-objects" can stream a slice of an existing packfile out when 254 the pack bitmap can tell that the reachable objects are all needed 255 in the output, without inspecting individual objects. This 256 strategy however would not work well when "--local" and other 257 options are in use, and need to be disabled. 258 259 * Fix memory leaks pointed out by Coverity (and people). 260 261 * "git read-tree -m" (no tree-ish) gave a nonsense suggestion "use 262 --empty if you want to clear the index". With "-m", such a request 263 will still fail anyway, as you'd need to name at least one tree-ish 264 to be merged. 265 266 * Make sure our tests would pass when the sources are checked out 267 with "platform native" line ending convention by default on 268 Windows. Some "text" files out tests use and the test scripts 269 themselves that are meant to be run with /bin/sh, ought to be 270 checked out with eol=LF even on Windows. 271 272 * Introduce the BUG() macro to improve die("BUG: ..."). 273 274 * Clarify documentation for include.path and includeIf.<condition>.path 275 configuration variables. 276 277 * Git sometimes gives an advice in a rhetorical question that does 278 not require an answer, which can confuse new users and non native 279 speakers. Attempt to rephrase them. 280 281 * A few http:// links that are redirected to https:// in the 282 documentation have been updated to https:// links. 283 284 * "git for-each-ref --format=..." with %(HEAD) in the format used to 285 resolve the HEAD symref as many times as it had processed refs, 286 which was wasteful, and "git branch" shared the same problem. 287 288 * Regression fix to topic recently merged to 'master'. 289 290 * The shell completion script (in contrib/) learned "git stash" has 291 a new "push" subcommand. 292 293 * "git interpret-trailers", when used as GIT_EDITOR for "git commit 294 -v", looked for and appended to a trailer block at the very end, 295 i.e. at the end of the "diff" output. The command has been 296 corrected to pay attention to the cut-mark line "commit -v" adds to 297 the buffer---the real trailer block should appear just before it. 298 299 * A test allowed both "git push" and "git receive-pack" on the other 300 end write their traces into the same file. This is OK on platforms 301 that allows atomically appending to a file opened with O_APPEND, 302 but on other platforms led to a mangled output, causing 303 intermittent test failures. This has been fixed by disabling 304 traces from "receive-pack" in the test. 305 306 * Tag objects, which are not reachable from any ref, that point at 307 missing objects were mishandled by "git gc" and friends (they 308 should silently be ignored instead) 309 310 * "git describe --contains" penalized light-weight tags so much that 311 they were almost never considered. Instead, give them about the 312 same chance to be considered as an annotated tag that is the same 313 age as the underlying commit would. 314 315 * The "run-command" API implementation has been made more robust 316 against dead-locking in a threaded environment. 317 318 * A recent update to t5545-push-options.sh started skipping all the 319 tests in the script when a web server testing is disabled or 320 unavailable, not just the ones that require a web server. Non HTTP 321 tests have been salvaged to always run in this script. 322 323 * "git send-email" now uses Net::SMTP::SSL, which is obsolete, only 324 when needed. Recent versions of Net::SMTP can do TLS natively. 325 326 * "foo\bar\baz" in "git fetch foo\bar\baz", even though there is no 327 slashes in it, cannot be a nickname for a remote on Windows, as 328 that is likely to be a pathname on a local filesystem. 329 330 * "git clean -d" used to clean directories that has ignored files, 331 even though the command should not lose ignored ones without "-x". 332 "git status --ignored" did not list ignored and untracked files 333 without "-uall". These have been corrected. 334 335 * The result from "git diff" that compares two blobs, e.g. "git diff 336 $commit1:$path $commit2:$path", used to be shown with the full 337 object name as given on the command line, but it is more natural to 338 use the $path in the output and use it to look up .gitattributes. 339 340 * The "collision detecting" SHA-1 implementation shipped with 2.13 341 was quite broken on some big-endian platforms and/or platforms that 342 do not like unaligned fetches. Update to the upstream code which 343 has already fixed these issues. 344 345 * "git am -h" triggered a BUG(). 346 347 * The interaction of "url.*.insteadOf" and custom URL scheme's 348 whitelisting is now documented better. 349 350 * The timestamp of the index file is now taken after the file is 351 closed, to help Windows, on which a stale timestamp is reported by 352 fstat() on a file that is opened for writing and data was written 353 but not yet closed. 354 355 * "git pull --rebase --autostash" didn't auto-stash when the local history 356 fast-forwards to the upstream. 357 358 * A flaky test has been corrected. 359 360 * "git $cmd -h" for builtin commands calls the implementation of the 361 command (i.e. cmd_$cmd() function) without doing any repository 362 set-up, and the commands that expect RUN_SETUP is done by the Git 363 potty needs to be prepared to show the help text without barfing. 364 (merge d691551192 jk/consistent-h later to maint). 365 366 * Help contributors that visit us at GitHub. 367 368 * "git stash push <pathspec>" did not work from a subdirectory at all. 369 Bugfix for a topic in v2.13 370 371 * As there is no portable way to pass timezone information to 372 strftime, some output format from "git log" and friends are 373 impossible to produce. Teach our own strbuf_addftime to replace %z 374 and %Z with caller-supplied values to help working around this. 375 (merge 6eced3ec5e rs/strbuf-addftime-zZ later to maint). 376 377 * "git mergetool" learned to work around a wrapper MacOS X adds 378 around underlying meld. 379 (merge 0af85f84bd da/mergetools-meld-output-opt-on-macos later to maint). 380 381 * An example in documentation that does not work in multi worktree 382 configuration has been corrected. 383 (merge 773a88914f ah/doc-gitattributes-empty-index later to maint). 384 385 * The pretty-format specifiers like '%h', '%t', etc. had an 386 optimization that no longer works correctly. In preparation/hope 387 of getting it correctly implemented, first discard the optimization 388 that is broken. 389 (merge fe9e2aefd4 rs/pretty-add-again later to maint). 390 391 * The code to pick up and execute command alias definition from the 392 configuration used to switch to the top of the working tree and 393 then come back when the expanded alias was executed, which was 394 unnecessarilyl complex. Attempt to simplify the logic by using the 395 early-config mechanism that does not chdir around. 396 (merge a9bcf6586d js/alias-early-config later to maint). 397 398 * Fix configuration codepath to pay proper attention to commondir 399 that is used in multi-worktree situation, and isolate config API 400 into its own header file. 401 (merge dc8441fdb4 bw/config-h later to maint). 402 403 * "git add -p" were updated in 2.12 timeframe to cope with custom 404 core.commentchar but the implementation was buggy and a 405 metacharacter like $ and * did not work. 406 (merge d85d7ecb80 jk/add-p-commentchar-fix later to maint). 407 408 * A recent regression in "git rebase -i" has been fixed and tests 409 that would have caught it and others have been added. 410 (merge adf16c08cb pw/rebase-i-regression-fix-tests later to maint). 411 412 * An unaligned 32-bit access in pack-bitmap code ahs been corrected. 413 (merge da41c942b3 jc/pack-bitmap-unaligned later to maint). 414 415 * Tighten error checks for invalid "git apply" input. 416 (merge d70e9c5c8c rs/apply-validate-input later to maint). 417 418 * Other minor doc, test and build updates and code cleanups. 419 (merge 68241cb9dd sb/t4005-modernize later to maint). 420 (merge 4fced24712 ks/t7508-indent-fix later to maint). 421 (merge 968b1fe263 mb/reword-autocomplete-message later to maint). 422 (merge 8592c95cdf ah/doc-pretty-color-auto-prefix later to maint).