1 Git v1.8.0 Release Notes
2 ========================
4 Backward compatibility notes
5 ----------------------------
7 In the next major release, we will change the behaviour of the "git
8 push" command. When "git push [$there]" does not say what to push, we
9 have used the traditional "matching" semantics (all your branches were
10 sent to the remote as long as there already are branches of the same
11 name over there). We will use the "simple" semantics, that pushes the
12 current branch to the branch with the same name only when the current
13 branch is set to integrate with that remote branch. There is a user
14 preference configuration variable "push.default" to change this, and
15 "git push" will warn about the upcoming change until you set this
18 "git branch --set-upstream" is deprecated and may be removed in a
19 relatively distant future. "git branch [-u|--set-upstream-to]" has
20 been introduced with a saner order of arguments.
26 UI, Workflows & Features
28 * A credential helper for Win32 to allow access to the keychain of
29 the logged-in user has been added.
31 * A credential helper to allow access to the Gnome keyring has been
34 * When "git am" sanitizes the Subject: line, we strip the prefix from
35 "Re: subject" and also from a less common "re: subject", but left
36 even less common "RE: subject" intact.
38 * It was tempting to say "git branch --set-upstream origin/master",
39 but that tells Git to arrange the local branch "origin/master" to
40 integrate with the currently checked out branch, which is highly
41 unlikely what the user meant. The option is deprecated; use the
42 new "--set-upstream-to" (with a short-and-sweet "-u") option
45 * "git cherry-pick" learned the "--allow-empty-message" option to
46 allow it to replay a commit without any log message.
48 * After "git cherry-pick -s" gave control back to the user asking
49 help to resolve conflicts, concluding "git commit" used to need to
50 be run with "-s" if the user wants to sign it off; now the command
51 leaves the sign-off line in the log template.
53 * "git daemon" learned the "--access-hook" option to allow an
54 external command to decline service based on the client address,
57 * "git difftool --dir-diff" learned to use symbolic links to prepare
58 temporary copy of the working tree when available.
60 * "git grep" learned to use a non-standard pattern type by default if
61 a configuration variable tells it to.
63 * "git merge-base" learned "--is-ancestor A B" option to tell if A is
64 an ancestor of B. The result is indicated by its exit status code.
66 * The "-Xours" backend option to "git merge -s recursive" now takes
67 effect even on binary files.
72 * "git svn" has been updated to work with SVN 1.7.
74 * "git p4" learned "--conflicts" option to specify what to do when
75 encountering a conflict during "p4 submit".
78 Performance, Internal Implementation, etc. (please report possible regressions)
80 * Git ships with a fall-back regexp implementation for platforms with
81 buggy regexp library, but it was easy for people to keep using their
82 platform regexp. A new test has been added to check this.
84 * The "check-docs" build target has been updated and greatly
87 * The documentation in the TeXinfo format was using indented output
88 for materials meant to be examples that are better typeset in
91 * Compatibility wrapper around some mkdir(2) implementations that
92 reject parameter with trailing slash has been introduced.
94 * Compatibility wrapper for systems that lack usable setitimer() has
97 * The option parsing of "git checkout" had error checking, dwim and
98 defaulting missing options, all mixed in the code, and issuing an
99 appropriate error message with useful context was getting harder.
100 The code has been reorganized to allow giving a proper diagnosis
101 when the user says "git checkout -b -t foo bar" (e.g. "-t" is not a
102 good name for a branch).
104 * Many internal uses of "git merge-base" equivalent were only to see
105 if one commit fast-forwards to the other, which did not need the
106 full set of merge bases to be computed. They have been updated to
107 use less expensive checks.
109 * The heuristics to detect and silently convert latin1 to utf8 when
110 we were told to use utf-8 in the log message has been transplanted
111 from "mailinfo" to "commit" and "commit-tree".
113 * Messages given by "git <subcommand> -h" from many subcommands have
114 been marked for translation.
117 Also contains minor documentation updates and code clean-ups.
123 Unless otherwise noted, all the fixes since v1.7.12 in the
124 maintenance track are contained in this release (see release notes
125 to them for details).
127 * "git blame MAKEFILE" run in a history that has "Makefile" but not
128 "MAKEFILE" should say "No such file MAKEFILE in HEAD", but got
129 confused on a case insensitive filesystem and failed to do so.
130 (merge 9aeaab6 jc/maint-blame-no-such-path later to maint).
132 * Even during a conflicted merge, "git blame $path" always meant to
133 blame uncommitted changes to the "working tree" version; make it
134 more useful by showing cleanly merged parts as coming from the other
135 branch that is being merged.
136 (merge 9aeaab6 jc/maint-blame-no-such-path later to maint).
138 * Documentation talked about "first line of commit log" when it meant
139 the title of the commit. The description was clarified by defining
140 how the title is decided and rewording the casual mention of "first
142 (merge 52ffe99 jw/doc-commit-title later to maint).
144 * The attribute system may be asked for a path that itself or its
145 leading directories no longer exists in the working tree, and it is
146 fine if we cannot open .gitattribute file in such a case. Failure
147 to open per-directory .gitattributes with error status other than
148 ENOENT and ENOTDIR should be diagnosed, but it wasn't.
150 * "git log --all-match --grep=A --grep=B" ought to show commits that
151 mention both A and B, but when these three options are used with
152 --author or --committer, it showed commits that mention either A or
154 (merge 39f2e01 jc/maint-log-grep-all-match later to maint).
156 * Earlier we made the diffstat summary line that shows the number of
157 lines added/deleted localizable, but it was found irritating having
158 to see them in various languages on a list whose discussion language
160 (merge 218adaa nd/maint-diffstat-summary later to maint).
162 * "git fetch --all", when passed "--no-tags", did not honor the
163 "--no-tags" option while fetching from individual remotes (the same
164 issue existed with "--tags", but combination "--all --tags" makes
165 much less sense than "--all --no-tags").
166 (merge 8556646 dj/fetch-all-tags later to maint).
168 * The subcommand in "git remote" to remove a defined remote was
169 "rm" and the command did not take a fully-spelled "remove".
170 (merge e17dba8 nd/maint-remote-remove later to maint).
172 * After "gitk" showed the contents of a tag, neither "Reread
173 references" nor "Reload" did not update what is shown as the
174 contents of it, when the user overwrote the tag with "git tag -f".
176 * "git cvsimport" did not thoroughly cleanse tag names that it
177 inferred from the names of the tags it obtained from CVS, which
178 caused "git tag" to barf and stop the import in the middle.
180 * "git show --format='%ci'" did not give timestamp correctly for
181 commits created without human readable name on "committer" line.
182 (merge e27ddb6 jc/maint-ident-missing-human-name later to maint).
184 * "git cherry-pick A C B" used to replay changes in A and then B and
185 then C if these three commits had committer timestamps in that
186 order, which is not what the user who said "A C B" naturally
188 (merge a73e22e mz/cherry-pick-cmdline-order later to maint).
190 * "git show --quiet" ought to be a synonym for "git show -s", but
192 (merge f9c75d8 jk/maint-quiet-is-synonym-to-s-in-log later to maint).
194 * "git p4", when "--use-client-spec" and "--detect-branches" are used
195 together, misdetected branches.
196 (merge 21ef5df pw/p4-use-client-spec-branch-detection later to maint).
198 * Output from "git branch -v" contains "(no branch)" that could be
199 localized, but the code to align it along with the names of
200 branches were counting in bytes, not in display columns.
201 (merge 1452bd6 nd/branch-v-alignment later to maint).
203 * When looking for $HOME/.gitconfig etc., it is OK if we cannot read
204 them because they do not exist, but we did not diagnose existing
205 files that we cannot read.
207 * The interactive prompt "git send-email" gives was error prone. It
208 asked "What e-mail address do you want to use?" with the address it
209 guessed (correctly) the user would want to use in its prompt,
210 tempting the user to say "y". But the response was taken as "No,
211 please use 'y' as the e-mail address instead", which is most
212 certainly not what the user meant.
213 (merge 6183749 sb/send-email-reconfirm-fix later to maint).