1Git v1.8.0 Release Notes 2======================== 3 4Backward compatibility notes 5---------------------------- 6 7In the next major release, we will change the behavior of the "git 8push" command. When "git push [$there]" does not say what to push, we 9have used the traditional "matching" semantics so far (all your branches were 10sent to the remote as long as there already are branches of the same 11name over there). We will now use the "simple" semantics, that pushes the 12current branch to the branch with the same name only when the current 13branch is set to integrate with that remote branch. There is a user 14preference configuration variable "push.default" to change this, and 15"git push" will warn about the upcoming change until you set this 16variable. 17 18"git branch --set-upstream" is deprecated and may be removed in a 19relatively distant future. "git branch [-u|--set-upstream-to]" has 20been introduced with a saner order of arguments. 21 22 23Updates since v1.7.12 24--------------------- 25 26UI, Workflows & Features 27 28 * A credential helper for Win32 to allow access to the keychain of 29 the logged-in user has been added. 30 31 * An initial port to HP NonStop. 32 33 * A credential helper to allow access to the Gnome keyring has been 34 added. 35 36 * When "git am" sanitizes the "Subject:" line, we strip the prefix from 37 "Re: subject" and also from a less common "re: subject", but left 38 the even less common "RE: subject" intact. We strip that now, too. 39 40 * It was tempting to say "git branch --set-upstream origin/master", 41 but that tells Git to arrange the local branch "origin/master" to 42 integrate with the currently checked out branch, which is highly 43 unlikely what the user meant. The option is deprecated; use the 44 new "--set-upstream-to" (with a short-and-sweet "-u") option 45 instead. 46 47 * "git cherry-pick" learned the "--allow-empty-message" option to 48 allow it to replay a commit without any log message. 49 50 * After "git cherry-pick -s" gave control back to the user asking 51 help to resolve conflicts, concluding "git commit" used to need to 52 be run with "-s" if the user wants to sign it off; now the command 53 leaves the sign-off line in the log template. 54 55 * "git daemon" learned the "--access-hook" option to allow an 56 external command to decline service based on the client address, 57 repository path, etc. 58 59 * "git difftool --dir-diff" learned to use symbolic links to prepare 60 a temporary copy of the working tree when available. 61 62 * "git grep" learned to use a non-standard pattern type by default if 63 a configuration variable tells it to. 64 65 * "git log -g" learned the "--grep-reflog=<pattern>" option to limit 66 its output to commits with a reflog message that matches the given 67 pattern. 68 69 * "git merge-base" learned the "--is-ancestor A B" option to tell if A is 70 an ancestor of B. The result is indicated by its exit status code. 71 72 * "git mergetool" now allows users to override the actual command used 73 with the mergetool.$name.cmd configuration variable even for built-in 74 mergetool backends. 75 76 * "git rebase -i" learned the "--edit-todo" option to open an editor 77 to edit the instruction sheet. 78 79 80Foreign Interface 81 82 * "git svn" has been updated to work with SVN 1.7. 83 84 * "git p4" learned the "--conflicts" option to specify what to do when 85 encountering a conflict during "p4 submit". 86 87 88Performance, Internal Implementation, etc. (please report possible regressions) 89 90 * Git ships with a fall-back regexp implementation for platforms with 91 buggy regexp library, but it was easy for people to keep using their 92 platform regexp by mistake. A new test has been added to check this. 93 94 * The "check-docs" build target has been updated and greatly 95 simplified. 96 97 * The test suite is run under MALLOC_CHECK_ when running with a glibc 98 that supports the feature. 99 100 * The documentation in the TeXinfo format was using indented output 101 for materials meant to be examples that are better typeset in 102 monospace. 103 104 * Compatibility wrapper around some mkdir(2) implementations that 105 reject parameters with trailing slash has been introduced. 106 107 * Compatibility wrapper for systems that lack usable setitimer() has 108 been added. 109 110 * The option parsing of "git checkout" had error checking, dwim and 111 defaulting missing options, all mixed in the code, and issuing an 112 appropriate error message with useful context was getting harder. 113 The code has been reorganized to allow giving a proper diagnosis 114 when the user says "git checkout -b -t foo bar" (e.g. "-t" is not a 115 good name for a branch). 116 117 * Many internal uses of a "git merge-base" equivalent were only to see 118 if one commit fast-forwards to the other, which did not need the 119 full set of merge bases to be computed. They have been updated to 120 use less expensive checks. 121 122 * The heuristics to detect and silently convert latin1 to utf8 when 123 we were told to use utf-8 in the log message has been transplanted 124 from "mailinfo" to "commit" and "commit-tree". 125 126 * Messages given by "git <subcommand> -h" from many subcommands have 127 been marked for translation. 128 129 130Also contains minor documentation updates and code clean-ups. 131 132 133Fixes since v1.7.12 134------------------- 135 136Unless otherwise noted, all the fixes since v1.7.12 in the 137maintenance track are contained in this release (see release notes 138to them for details). 139 140 * The attribute system may be asked for a path that itself or its 141 leading directories no longer exists in the working tree, and it is 142 fine if we cannot open .gitattribute file in such a case. Failure 143 to open per-directory .gitattributes with error status other than 144 ENOENT and ENOTDIR should be diagnosed, but it wasn't. 145 146 * When looking for $HOME/.gitconfig etc., it is OK if we cannot read 147 them because they do not exist, but we did not diagnose existing 148 files that we cannot read. 149 150 * When "git am" is fed an input that has multiple "Content-type: ..." 151 header, it did not grok charset= attribute correctly. 152 153 * "git am" mishandled a patch attached as application/octet-stream 154 (e.g. not text/*); Content-Transfer-Encoding (e.g. base64) was not 155 honored correctly. 156 157 * "git blame MAKEFILE" run in a history that has "Makefile" but not 158 "MAKEFILE" should say "No such file MAKEFILE in HEAD", but got 159 confused on a case insensitive filesystem and failed to do so. 160 161 * Even during a conflicted merge, "git blame $path" always meant to 162 blame uncommitted changes to the "working tree" version; make it 163 more useful by showing cleanly merged parts as coming from the other 164 branch that is being merged. 165 166 * It was unclear in the documentation for "git blame" that it is 167 unnecessary for users to use the "--follow" option. 168 169 * Output from "git branch -v" contains "(no branch)" that could be 170 localized, but the code to align it along with the names of 171 branches was counting in bytes, not in display columns. 172 173 * "git cherry-pick A C B" used to replay changes in A and then B and 174 then C if these three commits had committer timestamps in that 175 order, which is not what the user who said "A C B" naturally 176 expects. 177 178 * A repository created with "git clone --single" had its fetch 179 refspecs set up just like a clone without "--single", leading the 180 subsequent "git fetch" to slurp all the other branches, defeating 181 the whole point of specifying "only this branch". 182 183 * Documentation talked about "first line of commit log" when it meant 184 the title of the commit. The description was clarified by defining 185 how the title is decided and rewording the casual mention of "first 186 line" to "title". 187 188 * "git cvsimport" did not thoroughly cleanse tag names that it 189 inferred from the names of the tags it obtained from CVS, which 190 caused "git tag" to barf and stop the import in the middle. 191 192 * Earlier we made the diffstat summary line that shows the number of 193 lines added/deleted localizable, but it was found irritating having 194 to see them in various languages on a list whose discussion language 195 is English, and this change has been reverted. 196 197 * "git fetch --all", when passed "--no-tags", did not honor the 198 "--no-tags" option while fetching from individual remotes (the same 199 issue existed with "--tags", but the combination "--all --tags" makes 200 much less sense than "--all --no-tags"). 201 202 * "git fetch" over http had an old workaround for an unlikely server 203 misconfiguration; it turns out that this hurts debuggability of the 204 configuration in general, and has been reverted. 205 206 * "git fetch" over http advertised that it supports "deflate", which 207 is much less common, and did not advertise the more common "gzip" on 208 its Accept-Encoding header. 209 210 * "git fetch" over the dumb-http revision walker could segfault when 211 curl's multi interface was used. 212 213 * "git gc --auto" notified the user that auto-packing has triggered 214 even under the "--quiet" option. 215 216 * After "gitk" showed the contents of a tag, neither "Reread 217 references" nor "Reload" updated what is shown as the 218 contents of it when the user overwrote the tag with "git tag -f". 219 220 * "git log --all-match --grep=A --grep=B" ought to show commits that 221 mention both A and B, but when these three options are used with 222 --author or --committer, it showed commits that mention either A or 223 B (or both) instead. 224 225 * The "-Xours" backend option to "git merge -s recursive" was ignored 226 for binary files. 227 228 * "git p4", when "--use-client-spec" and "--detect-branches" are used 229 together, misdetected branches. 230 231 * "git receive-pack" (the counterpart to "git push") did not give 232 progress output while processing objects it received to the puser 233 when run over the smart-http protocol. 234 235 * When you misspell the command name you give to the "exec" action in 236 the "git rebase -i" instruction sheet you were told that 'rebase' is not a 237 git subcommand from "git rebase --continue". 238 239 * The subcommand in "git remote" to remove a defined remote was 240 "rm" and the command did not take a fully-spelled "remove". 241 242 * The interactive prompt that "git send-email" gives was error prone. It 243 asked "What e-mail address do you want to use?" with the address it 244 guessed (correctly) the user would want to use in its prompt, 245 tempting the user to say "y". But the response was taken as "No, 246 please use 'y' as the e-mail address instead", which is most 247 certainly not what the user meant. 248 249 * "git show --format='%ci'" did not give the timestamp correctly for 250 commits created without human readable name on the "committer" line. 251 252 * "git show --quiet" ought to be a synonym for "git show -s", but 253 wasn't. 254 255 * "git submodule frotz" was not diagnosed as "frotz" being an unknown 256 subcommand to "git submodule"; the user instead got a complaint 257 that "git submodule status" was run with an unknown path "frotz". 258 259 * "git status" honored the ignore=dirty settings in .gitmodules but 260 "git commit" didn't. 261 262 * "gitweb" did not give the correct committer timezone in its feed 263 output due to a typo.