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 (merge 867e40ff3a rs/large-zip later to maint). 42 43 * "git reset" learned "--recurse-submodules" option. 44 45 * "git diff --submodule=diff" now recurses into nested submodules. 46 (merge 5a5221427c jk/diff-submodule-diff-inline later to maint). 47 48 * "git repack" learned to accept the --threads=<n> option and pass it 49 to pack-objects. 50 51 * "git send-email" learned to run sendemail-validate hook to inspect 52 and reject a message before sending it out. 53 (merge 177409e589 jt/send-email-validate-hook later to maint). 54 55 * There is no good reason why "git fetch $there $sha1" should fail 56 when the $sha1 names an object at the tip of an advertised ref, 57 even when the other side hasn't enabled allowTipSHA1InWant. 58 59 * The recently introduced "[includeIf "gitdir:$dir"] path=..." 60 mechansim has further been taught to take symlinks into account. 61 The directory "$dir" specified in "gitdir:$dir" may be a symlink to 62 a real location, not something that $(getcwd) may return. In such 63 a case, a realpath of "$dir" is compared with the real path of the 64 current repository to determine if the contents from the named path 65 should be included. 66 67 * Make the "indent" heuristics the default in "diff" and diff.indentHeuristics 68 configuration variable an escape hatch for those who do no want it. 69 70 * Many commands learned to pay attention to submodule.recurse 71 configuration. 72 73 * The convention for a command line is to follow "git cmdname 74 --options" with revisions followed by an optional "--" 75 disambiguator and then finally pathspecs. When "--" is not there, 76 we make sure early ones are all interpretable as revs (and do not 77 look like paths) and later ones are the other way around. A 78 pathspec with "magic" (e.g. ":/p/a/t/h" that matches p/a/t/h from 79 the top-level of the working tree, no matter what subdirectory you 80 are working from) are conservatively judged as "not a path", which 81 required disambiguation more often. The command line parser 82 learned to say "it's a pathspec" a bit more often when the syntax 83 looks like so. 84 (merge 2cb47ab695 jk/pathspec-magic-disambiguation later to maint). 85 86 * Update "perl-compatible regular expression" support to enable JIT 87 and also allow linking with the newer PCRE v2 library. 88 89 90Performance, Internal Implementation, Development Support etc. 91 92 * The default packed-git limit value has been raised on larger 93 platforms to save "git fetch" from a (recoverable) failure while 94 "gc" is running in parallel. 95 96 * Code to update the cache-tree has been tightened so that we won't 97 accidentally write out any 0{40} entry in the tree object. 98 (merge a96d3cc3f6 jk/no-null-sha1-in-cache-tree later to maint). 99 100 * Attempt to allow us notice "fishy" situation where we fail to 101 remove the temporary directory used during the test. 102 103 * Travis CI gained a task to format the documentation with both 104 AsciiDoc and AsciiDoctor. 105 106 * Some platforms have ulong that is smaller than time_t, and our 107 historical use of ulong for timestamp would mean they cannot 108 represent some timestamp that the platform allows. Invent a 109 separate and dedicated timestamp_t (so that we can distingiuish 110 timestamps and a vanilla ulongs, which along is already a good 111 move), and then declare uintmax_t is the type to be used as the 112 timestamp_t. 113 114 * We can trigger Windows auto-build tester (credits: Dscho & 115 Microsoft) from our existing Travis CI tester now. 116 117 * Conversion from uchar[20] to struct object_id continues. 118 119 * Simplify parse_pathspec() codepath and stop it from looking at the 120 default in-core index. 121 (merge 08de9151a8 bw/pathspec-sans-the-index later to maint). 122 123 * Add perf-test for wildmatch. 124 (merge 62ca75a6b9 ab/perf-wildmatch later to maint). 125 126 * Code from "conversion using external process" codepath has been 127 extracted to a separate sub-process.[ch] module. 128 (merge 4f2a2e9f0e bp/sub-process-convert-filter later to maint). 129 130 * When "git checkout", "git merge", etc. manipulates the in-core 131 index, various pieces of information in the index extensions are 132 discarded from the original state, as it is usually not the case 133 that they are kept up-to-date and in-sync with the operation on the 134 main index. The untracked cache extension is copied across these 135 operations now, which would speed up "git status" (as long as the 136 cache is properly invalidated). 137 138 * The internal implementation of "git grep" has seen some clean-up. 139 (merge 8df4c2953f ab/grep-preparatory-cleanup later to maint). 140 141 * Update the C style recommendation for notes for translators, as 142 recent versions of gettext tools can work with our style of 143 multi-line comments. 144 145 * The implementation of "ref" API around the "packed refs" have been 146 cleaned up, in preparation for further changes. 147 148 * The internal logic used in "git blame" has been libified to make it 149 easier to use by cgit. 150 151 * Our code often opens a path to an optional file, to work on its 152 contents when we can successfully open it. We can ignore a failure 153 to open if such an optional file does not exist, but we do want to 154 report a failure in opening for other reasons (e.g. we got an I/O 155 error, or the file is there, but we lack the permission to open). 156 157 The exact errors we need to ignore are ENOENT (obviously) and 158 ENOTDIR (less obvious). Instead of repeating comparison of errno 159 with these two constants, introduce a helper function to do so. 160 161 * We often try to open a file for reading whose existence is 162 optional, and silently ignore errors from open/fopen; report such 163 errors if they are not due to missing files. 164 165 * When an existing repository is used for t/perf testing, we first 166 create bit-for-bit copy of it, which may grab a transient state of 167 the repository and freeze it into the repository used for testing, 168 which then may cause Git operations to fail. Single out "the index 169 being locked" case and forcibly drop the lock from the copy. 170 171 172Also contains various documentation updates and code clean-ups. 173 174 175Fixes since v2.13 176----------------- 177 178Unless otherwise noted, all the fixes since v2.13 in the maintenance 179track are contained in this release (see the maintenance releases' 180notes for details). 181 182 * "git gc" did not interact well with "git worktree"-managed 183 per-worktree refs. 184 185 * "git cherry-pick" and other uses of the sequencer machinery 186 mishandled a trailer block whose last line is an incomplete line. 187 This has been fixed so that an additional sign-off etc. are added 188 after completing the existing incomplete line. 189 190 * The codepath in "git am" that is used when running "git rebase" 191 leaked memory held for the log message of the commits being rebased. 192 193 * "git clone --config var=val" is a way to populate the 194 per-repository configuration file of the new repository, but it did 195 not work well when val is an empty string. This has been fixed. 196 197 * Setting "log.decorate=false" in the configuration file did not take 198 effect in v2.13, which has been corrected. 199 200 * A few codepaths in "checkout" and "am" working on an unborn branch 201 tried to access an uninitialized piece of memory. 202 203 * The Web interface to gmane news archive is long gone, even though 204 the articles are still accessible via NTTP. Replace the links with 205 ones to public-inbox.org. Because their message identification is 206 based on the actual message-id, it is likely that it will be easier 207 to migrate away from it if/when necessary. 208 209 * The receive-pack program now makes sure that the push certificate 210 records the same set of push options used for pushing. 211 212 * Tests have been updated to pass under GETTEXT_POISON (a mechanism 213 to ensure that output strings that should not be translated are 214 not translated by mistake), and TravisCI is told to run them. 215 216 * "git checkout --recurse-submodules" did not quite work with a 217 submodule that itself has submodules. 218 219 * Plug some leaks and updates internal API used to implement the 220 split index feature to make it easier to avoid such a leak in the 221 future. 222 (merge de950c5773 nd/split-index-unshare later to maint). 223 224 * "pack-objects" can stream a slice of an existing packfile out when 225 the pack bitmap can tell that the reachable objects are all needed 226 in the output, without inspecting individual objects. This 227 strategy however would not work well when "--local" and other 228 options are in use, and need to be disabled. 229 230 * Fix memory leaks pointed out by Coverity (and people). 231 232 * "git read-tree -m" (no tree-ish) gave a nonsense suggestion "use 233 --empty if you want to clear the index". With "-m", such a request 234 will still fail anyway, as you'd need to name at least one tree-ish 235 to be merged. 236 237 * Make sure our tests would pass when the sources are checked out 238 with "platform native" line ending convention by default on 239 Windows. Some "text" files out tests use and the test scripts 240 themselves that are meant to be run with /bin/sh, ought to be 241 checked out with eol=LF even on Windows. 242 243 * Introduce the BUG() macro to improve die("BUG: ..."). 244 245 * Clarify documentation for include.path and includeIf.<condition>.path 246 configuration variables. 247 248 * Git sometimes gives an advice in a rhetorical question that does 249 not require an answer, which can confuse new users and non native 250 speakers. Attempt to rephrase them. 251 252 * A few http:// links that are redirected to https:// in the 253 documentation have been updated to https:// links. 254 255 * "git for-each-ref --format=..." with %(HEAD) in the format used to 256 resolve the HEAD symref as many times as it had processed refs, 257 which was wasteful, and "git branch" shared the same problem. 258 259 * Regression fix to topic recently merged to 'master'. 260 261 * The shell completion script (in contrib/) learned "git stash" has 262 a new "push" subcommand. 263 264 * "git interpret-trailers", when used as GIT_EDITOR for "git commit 265 -v", looked for and appended to a trailer block at the very end, 266 i.e. at the end of the "diff" output. The command has been 267 corrected to pay attention to the cut-mark line "commit -v" adds to 268 the buffer---the real trailer block should appear just before it. 269 270 * A test allowed both "git push" and "git receive-pack" on the other 271 end write their traces into the same file. This is OK on platforms 272 that allows atomically appending to a file opened with O_APPEND, 273 but on other platforms led to a mangled output, causing 274 intermittent test failures. This has been fixed by disabling 275 traces from "receive-pack" in the test. 276 277 * Tag objects, which are not reachable from any ref, that point at 278 missing objects were mishandled by "git gc" and friends (they 279 should silently be ignored instead) 280 281 * "git describe --contains" penalized light-weight tags so much that 282 they were almost never considered. Instead, give them about the 283 same chance to be considered as an annotated tag that is the same 284 age as the underlying commit would. 285 (merge ef1e74065c jc/name-rev-lw-tag later to maint). 286 287 * The "run-command" API implementation has been made more robust 288 against dead-locking in a threaded environment. 289 290 * A recent update to t5545-push-options.sh started skipping all the 291 tests in the script when a web server testing is disabled or 292 unavailable, not just the ones that require a web server. Non HTTP 293 tests have been salvaged to always run in this script. 294 295 * "git send-email" now uses Net::SMTP::SSL, which is obsolete, only 296 when needed. Recent versions of Net::SMTP can do TLS natively. 297 298 * "foo\bar\baz" in "git fetch foo\bar\baz", even though there is no 299 slashes in it, cannot be a nickname for a remote on Windows, as 300 that is likely to be a pathname on a local filesystem. 301 302 * "git clean -d" used to clean directories that has ignored files, 303 even though the command should not lose ignored ones without "-x". 304 "git status --ignored" did not list ignored and untracked files 305 without "-uall". These have been corrected. 306 307 * The result from "git diff" that compares two blobs, e.g. "git diff 308 $commit1:$path $commit2:$path", used to be shown with the full 309 object name as given on the command line, but it is more natural to 310 use the $path in the output and use it to look up .gitattributes. 311 (merge 30d005c020 jk/diff-blob later to maint). 312 313 * The "collision detecting" SHA-1 implementation shipped with 2.13 314 was quite broken on some big-endian platforms and/or platforms that 315 do not like unaligned fetches. Update to the upstream code which 316 has already fixed these issues. 317 318 * "git am -h" triggered a BUG(). 319 320 * The interaction of "url.*.insteadOf" and custom URL scheme's 321 whitelisting is now documented better. 322 323 * The timestamp of the index file is now taken after the file is 324 closed, to help Windows, on which a stale timestamp is reported by 325 fstat() on a file that is opened for writing and data was written 326 but not yet closed. 327 328 * "git pull --rebase --autostash" didn't auto-stash when the local history 329 fast-forwards to the upstream. 330 331 * A flaky test has been corrected. 332 (merge 7c2115aa07 jk/pack-idx-corruption-safety later to maint). 333 334 * "git $cmd -h" for builtin commands calls the implementation of the 335 command (i.e. cmd_$cmd() function) without doing any repository 336 set-up, and the commands that expect RUN_SETUP is done by the Git 337 potty needs to be prepared to show the help text without barfing. 338 (merge d691551192 jk/consistent-h later to maint). 339 340 * Other minor doc, test and build updates and code cleanups. 341 (merge 8ba74bfd7c jc/diff-tree-stale-comment later to maint). 342 (merge 68602c01fd sb/submodule-rm-absorb later to maint).