1Git 2.15 Release Notes 2====================== 3 4Backward compatibility notes and other notable changes. 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 now scheduled to happen in the upcoming 12 release. 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 * "branch --set-upstream" that has been deprecated in Git 1.8 has 22 finally been retired. 23 24 25Updates since v2.14 26------------------- 27 28UI, Workflows & Features 29 30 * An example that is now obsolete has been removed from a sample hook, 31 and an old example in it that added a sign-off manually has been 32 improved to use the interpret-trailers command. 33 34 * The advice message given when "git rebase" stops for conflicting 35 changes has been improved. 36 37 * The "rerere-train" script (in contrib/) learned the "--overwrite" 38 option to allow overwriting existing recorded resolutions. 39 40 * "git contacts" (in contrib/) now lists the address on the 41 "Reported-by:" trailer to its output, in addition to those on 42 S-o-b: and other trailers, to make it easier to notify (and thank) 43 the original bug reporter. 44 45 * "git rebase", especially when it is run by mistake and ends up 46 trying to replay many changes, spent long time in silence. The 47 command has been taught to show progress report when it spends 48 long time preparing these many changes to replay (which would give 49 the user a chance to abort with ^C). 50 51 * "git merge" learned a "--signoff" option to add the Signed-off-by: 52 trailer with the committer's name. 53 54 * "git diff" learned to optionally paint new lines that are the same 55 as deleted lines elsewhere differently from genuinely new lines. 56 57 * "git interpret-trailers" learned to take the trailer specifications 58 from the command line that overrides the configured values. 59 60 * "git interpret-trailers" has been taught a "--parse" and a few 61 other options to make it easier for scripts to grab existing 62 trailer lines from a commit log message. 63 64 * "gitweb" shows a link to visit the 'raw' contents of blbos in the 65 history overview page. 66 67 * "[gc] rerereResolved = 5.days" used to be invalid, as the variable 68 is defined to take an integer counting the number of days. It now 69 is allowed. 70 71 * The code to acquire a lock on a reference (e.g. while accepting a 72 push from a client) used to immediately fail when the reference is 73 already locked---now it waits for a very short while and retries, 74 which can make it succeed if the lock holder was holding it during 75 a read-only operation. 76 77 * "branch --set-upstream" that has been deprecated in Git 1.8 has 78 finally been retired. 79 80 * The codepath to call external process filter for smudge/clean 81 operation learned to show the progress meter. 82 83 84Performance, Internal Implementation, Development Support etc. 85 86 * Conversion from uchar[20] to struct object_id continues. 87 88 * Start using selected c99 constructs in small, stable and 89 essentialpart of the system to catch people who care about 90 older compilers that do not grok them. 91 92 * The filter-process interface learned to allow a process with long 93 latency give a "delayed" response. 94 95 * Many uses of comparision callback function the hashmap API uses 96 cast the callback function type when registering it to 97 hashmap_init(), which defeats the compile time type checking when 98 the callback interface changes (e.g. gaining more parameters). 99 The callback implementations have been updated to take "void *" 100 pointers and cast them to the type they expect instead. 101 102 * Because recent Git for Windows do come with a real msgfmt, the 103 build procedure for git-gui has been updated to use it instead of a 104 hand-rolled substitute. 105 106 * "git grep --recurse-submodules" has been reworked to give a more 107 consistent output across submodule boundary (and do its thing 108 without having to fork a separate process). 109 110 * A helper function to read a single whole line into strbuf 111 mistakenly triggered OOM error at EOF under certain conditions, 112 which has been fixed. 113 (merge 642956cf45 rs/strbuf-getwholeline-fix later to maint). 114 115 * The "ref-store" code reorganization continues. 116 117 * "git commit" used to discard the index and re-read from the filesystem 118 just in case the pre-commit hook has updated it in the middle; this 119 has been optimized out when we know we do not run the pre-commit hook. 120 (merge 680ee550d7 kw/commit-keep-index-when-pre-commit-is-not-run later to maint). 121 122 * Updates to the HTTP layer we made recently unconditionally used 123 features of libCurl without checking the existence of them, causing 124 compilation errors, which has been fixed. Also migrate the code to 125 check feature macros, not version numbers, to cope better with 126 libCurl that vendor ships with backported features. 127 128 * The API to start showing progress meter after a short delay has 129 been simplified. 130 (merge 8aade107dd jc/simplify-progress later to maint). 131 132 * Code clean-up to avoid mixing values read from the .gitmodules file 133 and values read from the .git/config file. 134 135 * We used to spend more than necessary cycles allocating and freeing 136 piece of memory while writing each index entry out. This has been 137 optimized. 138 139 * Platforms that ship with a separate sha1 with collision detection 140 library can link to it instead of using the copy we ship as part of 141 our source tree. 142 143 * Code around "notes" have been cleaned up. 144 (merge 3964281524 mh/notes-cleanup later to maint). 145 146 * The long-standing rule that an in-core lockfile instance, once it 147 is used, must not be freed, has been lifted and the lockfile and 148 tempfile APIs have been updated to reduce the chance of programming 149 errors. 150 151 * Our hashmap implementation in hashmap.[ch] is not thread-safe when 152 adding a new item needs to expand the hashtable by rehashing; add 153 an API to disable the automatic rehashing to work it around. 154 155 * Many of our programs consider that it is OK to release dynamic 156 storage that is used throughout the life of the program by simply 157 exiting, but this makes it harder to leak detection tools to avoid 158 reporting false positives. Plug many existing leaks and introduce 159 a mechanism for developers to mark that the region of memory 160 pointed by a pointer is not lost/leaking to help these tools. 161 162 * As "git commit" to conclude a conflicted "git merge" honors the 163 commit-msg hook, "git merge" that records a merge commit that 164 cleanly auto-merges should, but it didn't. 165 166 * The codepath for "git merge-recursive" has been cleaned up. 167 168 * Many leaks of strbuf have been fixed. 169 170 171Also contains various documentation updates and code clean-ups. 172 173 174Fixes since v2.14 175----------------- 176 177 * "%C(color name)" in the pretty print format always produced ANSI 178 color escape codes, which was an early design mistake. They now 179 honor the configuration (e.g. "color.ui = never") and also tty-ness 180 of the output medium. 181 182 * The http.{sslkey,sslCert} configuration variables are to be 183 interpreted as a pathname that honors "~[username]/" prefix, but 184 weren't, which has been fixed. 185 186 * Numerous bugs in walking of reflogs via "log -g" and friends have 187 been fixed. 188 189 * "git commit" when seeing an totally empty message said "you did not 190 edit the message", which is clearly wrong. The message has been 191 corrected. 192 193 * When a directory is not readable, "gitweb" fails to build the 194 project list. Work this around by skipping such a directory. 195 196 * Some versions of GnuPG fails to kill gpg-agent it auto-spawned 197 and such a left-over agent can interfere with a test. Work it 198 around by attempting to kill one before starting a new test. 199 200 * A recently added test for the "credential-cache" helper revealed 201 that EOF detection done around the time the connection to the cache 202 daemon is torn down were flaky. This was fixed by reacting to 203 ECONNRESET and behaving as if we got an EOF. 204 205 * "git log --tag=no-such-tag" showed log starting from HEAD, which 206 has been fixed---it now shows nothing. 207 208 * The "tag.pager" configuration variable was useless for those who 209 actually create tag objects, as it interfered with the use of an 210 editor. A new mechanism has been introduced for commands to enable 211 pager depending on what operation is being carried out to fix this, 212 and then "git tag -l" is made to run pager by default. 213 214 * "git push --recurse-submodules $there HEAD:$target" was not 215 propagated down to the submodules, but now it is. 216 217 * Commands like "git rebase" accepted the --rerere-autoupdate option 218 from the command line, but did not always use it. This has been 219 fixed. 220 221 * "git clone --recurse-submodules --quiet" did not pass the quiet 222 option down to submodules. 223 224 * Test portability fix for OBSD. 225 226 * Portability fix for OBSD. 227 228 * "git am -s" has been taught that some input may end with a trailer 229 block that is not Signed-off-by: and it should refrain from adding 230 an extra blank line before adding a new sign-off in such a case. 231 232 * "git svn" used with "--localtime" option did not compute the tz 233 offset for the timestamp in question and instead always used the 234 current time, which has been corrected. 235 236 * Memory leak in an error codepath has been plugged. 237 238 * "git stash -u" used the contents of the committed version of the 239 ".gitignore" file to decide which paths are ignored, even when the 240 file has local changes. The command has been taught to instead use 241 the locally modified contents. 242 243 * bash 4.4 or newer gave a warning on NUL byte in command 244 substitution done in "git stash"; this has been squelched. 245 246 * "git grep -L" and "git grep --quiet -L" reported different exit 247 codes; this has been corrected. 248 249 * When handshake with a subprocess filter notices that the process 250 asked for an unknown capability, Git did not report what program 251 the offending subprocess was running. This has been corrected. 252 253 * "git apply" that is used as a better "patch -p1" failed to apply a 254 taken from a file with CRLF line endings to a file with CRLF line 255 endings. The root cause was because it misused convert_to_git() 256 that tried to do "safe-crlf" processing by looking at the index 257 entry at the same path, which is a nonsense---in that mode, "apply" 258 is not working on the data in (or derived from) the index at all. 259 This has been fixed. 260 261 * Killing "git merge --edit" before the editor returns control left 262 the repository in a state with MERGE_MSG but without MERGE_HEAD, 263 which incorrectly tells the subsequent "git commit" that there was 264 a squash merge in progress. This has been fixed. 265 266 * "git archive" did not work well with pathspecs and the 267 export-ignore attribute. 268 269 * In addition to "cc: <a@dd.re.ss> # cruft", "cc: a@dd.re.ss # cruft" 270 was taught to "git send-email" as a valid way to tell it that it 271 needs to also send a carbon copy to <a@dd.re.ss> in the trailer 272 section. 273 (merge cc90750677 mm/send-email-cc-cruft later to maint). 274 275 * "git branch -M a b" while on a branch that is completely unrelated 276 to either branch a or branch b misbehaved when multiple worktree 277 was in use. This has been fixed. 278 (merge 31824d180d nd/worktree-kill-parse-ref later to maint). 279 280 * "git gc" and friends when multiple worktrees are used off of a 281 single repository did not consider the index and per-worktree refs 282 of other worktrees as the root for reachability traversal, making 283 objects that are in use only in other worktrees to be subject to 284 garbage collection. 285 286 * A regression to "gitk --bisect" by a recent update has been fixed. 287 (merge 1d0538e486 mh/packed-ref-store-prep later to maint). 288 289 * "git -c submodule.recurse=yes pull" did not work as if the 290 "--recurse-submodules" option was given from the command line. 291 This has been corrected. 292 293 * Unlike "git commit-tree < file", "git commit-tree -F file" did not 294 pass the contents of the file verbatim and instead completed an 295 incomplete line at the end, if exists. The latter has been updated 296 to match the behaviour of the former. 297 (merge c818e74332 rk/commit-tree-make-F-verbatim later to maint). 298 299 * Other minor doc, test and build updates and code cleanups. 300 (merge f094b89a4d ma/parse-maybe-bool later to maint). 301 (merge 39b00fa4d4 jk/drop-sha1-entry-pos later to maint). 302 (merge 6cdf8a7929 ma/ts-cleanups later to maint). 303 (merge 7560f547e6 ma/up-to-date later to maint). 304 (merge 0db3dc75f3 rs/apply-epoch later to maint). 305 (merge 74f1bd912b dw/diff-highlight-makefile-fix later to maint). 306 (merge f991761eb8 jk/config-lockfile-leak-fix later to maint). 307 (merge 150efef1e7 ma/pkt-line-leakfix later to maint). 308 (merge 5554451de6 mg/timestamp-t-fix later to maint). 309 (merge 276d0e35c0 ma/split-symref-update-fix later to maint). 310 (merge 3bc4b8f7c7 bb/doc-eol-dirty later to maint). 311 (merge c1bb33c99c jk/system-path-cleanup later to maint).