| blob | f335bcfcdd8a32ab323630bafed25829177d1ed3 |
1 Git v1.8.5 Release Notes
2 ========================
4 Backward compatibility notes (for Git 2.0)
5 ------------------------------------------
7 When "git push [$there]" does not say what to push, we have used the
8 traditional "matching" semantics so far (all your branches were sent
9 to the remote as long as there already are branches of the same name
10 over there). In Git 2.0, the default will change to the "simple"
11 semantics that pushes:
13 - only the current branch to the branch with the same name, and only
14 when the current branch is set to integrate with that remote
15 branch, if you are pushing to the same remote as you fetch from; or
17 - only the current branch to the branch with the same name, if you
18 are pushing to a remote that is not where you usually fetch from.
20 Use the user preference configuration variable "push.default" to
21 change this. If you are an old-timer who is used to the "matching"
22 semantics, you can set the variable to "matching" to keep the
23 traditional behaviour. If you want to live in the future early, you
24 can set it to "simple" today without waiting for Git 2.0.
26 When "git add -u" (and "git add -A") is run inside a subdirectory and
27 does not specify which paths to add on the command line, it
28 will operate on the entire tree in Git 2.0 for consistency
29 with "git commit -a" and other commands. There will be no
30 mechanism to make plain "git add -u" behave like "git add -u .".
31 Current users of "git add -u" (without a pathspec) should start
32 training their fingers to explicitly say "git add -u ."
33 before Git 2.0 comes. A warning is issued when these commands are
34 run without a pathspec and when you have local changes outside the
35 current directory, because the behaviour in Git 2.0 will be different
36 from today's version in such a situation.
38 In Git 2.0, "git add <path>" will behave as "git add -A <path>", so
39 that "git add dir/" will notice paths you removed from the directory
40 and record the removal. Versions before Git 2.0, including this
41 release, will keep ignoring removals, but the users who rely on this
42 behaviour are encouraged to start using "git add --ignore-removal <path>"
43 now before 2.0 is released.
46 Updates since v1.8.4
47 --------------------
49 Foreign interfaces, subsystems and ports.
51 * On MacOS X, we detected if the filesystem needs the "pre-composed
52 unicode strings" workaround, but did not automatically enable it.
53 Now we do.
55 * remote-hg remote helper misbehaved when interacting with a local Hg
56 repository relative to the home directory, e.g. "clone hg::~/there".
58 * imap-send ported to OS X uses Apple's security framework instead of
59 OpenSSL one.
61 * Subversion 1.8.0 that was recently released breaks older subversion
62 clients coming over http/https in various ways.
64 * "git fast-import" treats an empty path given to "ls" as the root of
65 the tree.
68 UI, Workflows & Features
70 * Earlier we started rejecting an attempt to add 0{40} object name to
71 the index and to tree objects, but it sometimes is necessary to
72 allow so to be able to use tools like filter-branch to correct such
73 broken tree objects. "filter-branch" can again be used to to do
74 so.
76 * "git config" did not provide a way to set or access numbers larger
77 than a native "int" on the platform; it now provides 64-bit signed
78 integers on all platforms.
80 * "git pull --rebase" always chose to do the bog-standard flattening
81 rebase. You can tell it to run "rebase --preserve-merges" by
82 setting "pull.rebase" configuration to "preserve".
84 * "git push --no-thin" actually disables the "thin pack transfer"
85 optimization.
87 * Magic pathspecs like ":(icase)makefile" that matches both
88 Makefile and makefile can be used in more places.
90 * The "http.*" variables can now be specified per URL that the
91 configuration applies. For example,
93 [http]
94 sslVerify = true
95 [http "https://weak.example.com/"]
96 sslVerify = false
98 would flip http.sslVerify off only when talking to that specified
99 site.
101 * "git mv A B" when moving a submodule A has been taught to
102 relocate its working tree and to adjust the paths in the
103 .gitmodules file.
105 * "git blame" can now take more than one -L option to discover the
106 origin of multiple blocks of the lines.
108 * The http transport clients can optionally ask to save cookies
109 with http.savecookies configuration variable.
111 * "git push" learned a more fine grained control over a blunt
112 "--force" when requesting a non-fast-forward update with the
113 "--force-with-lease=<refname>:<expected object name>" option.
115 * "git diff --diff-filter=<classes of changes>" can now take
116 lowercase letters (e.g. "--diff-filter=d") to mean "show
117 everything but these classes". "git diff-files -q" is now a
118 deprecated synonym for "git diff-files --diff-filter=d".
120 * "git fetch" (hence "git pull" as well) learned to check
121 "fetch.prune" and "remote.*.prune" configuration variables and
122 to behave as if the "--prune" command line option was given.
124 * "git check-ignore -z" applied the NUL termination to both its input
125 (with --stdin) and its output, but "git check-attr -z" ignored the
126 option on the output side. Make both honor -z on the input and
127 output side the same way.
129 * "git whatchanged" may still be used by old timers, but mention of
130 it in documents meant for new users will only waste readers' time
131 wonderig what the difference is between it and "git log". Make it
132 less prominent in the general part of the documentation and explain
133 that it is merely a "git log" with different default behaviour in
134 its own document.
137 Performance, Internal Implementation, etc.
139 * Many commands use --dashed-option as a operation mode selector
140 (e.g. "git tag --delete") that the user can use at most one
141 (e.g. "git tag --delete --verify" is a nonsense) and you cannot
142 negate (e.g. "git tag --no-delete" is a nonsense). parse-options
143 API learned a new OPT_CMDMODE macro to make it easier to implement
144 such a set of options.
146 * OPT_BOOLEAN() in parse-options API was misdesigned to be "counting
147 up" but many subcommands expect it to behave as "on/off". Update
148 them to use OPT_BOOL() which is a proper boolean.
150 * "git gc" exits early without doing a double-work when it detects
151 that another instance of itself is already running.
153 * Under memory pressure and/or file descriptor pressure, we used to
154 close pack windows that are not used and also closed filehandle to
155 an open but unused packfiles. These are now controlled separately
156 to better cope with the load.
158 Also contains various documentation updates and code clean-ups.
161 Fixes since v1.8.4
162 ------------------
164 Unless otherwise noted, all the fixes since v1.8.4 in the maintenance
165 track are contained in this release (see release notes to them for
166 details).
168 * When an object is not found after checking the packfiles and then
169 loose object directory, read_sha1_file() re-checks the packfiles to
170 prevent racing with a concurrent repacker; teach the same logic to
171 has_sha1_file().
172 (merge 45e8a74 jk/has-sha1-file-retry-packed later to maint).
174 * "git commit --author=$name", when $name is not in the canonical
175 "A. U. Thor <au.thor@example.xz>" format, looks for a matching name
176 from existing history, but did not consult mailmap to grab the
177 preferred author name.
178 (merge ea16794 ap/commit-author-mailmap later to maint).
180 * "git ls-files -k" needs to crawl only the part of the working tree
181 that may overlap the paths in the index to find killed files, but
182 shared code with the logic to find all the untracked files, which
183 made it unnecessarily inefficient.
184 (merge 680be04 jc/ls-files-killed-optim later to maint).
186 * The commit object names in the insn sheet that was prepared at the
187 beginning of "rebase -i" session can become ambiguous as the
188 rebasing progresses and the repository gains more commits. Make
189 sure the internal record is kept with full 40-hex object names.
190 (merge 75c6976 es/rebase-i-no-abbrev later to maint).
192 * "git rebase --preserve-merges" internally used the merge machinery
193 and as a side effect, left merge summary message in the log, but
194 when rebasing, there should not be a need for merge summary.
195 (merge a9f739c rt/rebase-p-no-merge-summary later to maint).
197 * A call to xread() was used without a loop around to cope with short
198 read in the codepath to stream new contents to a pack.
199 (merge e92527c js/xread-in-full later to maint).
201 * "git rebase -i" forgot that the comment character can be
202 configurable while reading its insn sheet.
203 (merge 7bca7af es/rebase-i-respect-core-commentchar later to maint).
205 * The mailmap support code read past the allocated buffer when the
206 mailmap file ended with an incomplete line.
207 (merge f972a16 jk/mailmap-incomplete-line later to maint).
209 * We used to send a large request to read(2)/write(2) as a single
210 system call, which was bad from the latency point of view when
211 the operation needs to be killed, and also triggered an error on
212 broken 64-bit systems that refuse to take more than 2GB read or
213 write in one go.
214 (merge a487916 sp/clip-read-write-to-8mb later to maint).
216 * "git fetch" that auto-followed tags incorrectly reused the
217 connection with Git-aware transport helper (like the sample "ext::"
218 helper shipped with Git).
219 (merge 0f73f8b jc/transport-do-not-use-connect-twice-in-fetch later to maint).
221 * "git log --full-diff -- <pathspec>" showed a huge diff for paths
222 outside the given <pathspec> for each commit, instead of showing
223 the change relative to the parent of the commit. "git reflog -p"
224 had a similar problem.
225 (merge 838f9a1 tr/log-full-diff-keep-true-parents later to maint).
227 * Setting submodule.*.path configuration variable to true (without
228 giving "= value") caused Git to segfault.
229 (merge 4b05440 jl/some-submodule-config-are-not-boolean later to maint).
231 * "git rebase -i" (there could be others, as the root cause is pretty
232 generic) fed a random, data dependeant string to 'echo' and
233 expects it to come out literally, corrupting its error message.
234 (merge 89b0230 mm/no-shell-escape-in-die-message later to maint).
236 * Some people still use rather old versions of bash, which cannot
237 grok some constructs like 'printf -v varname' the prompt and
238 completion code started to use recently.
239 (merge a44aa69 bc/completion-for-bash-3.0 later to maint).
241 * Code to read configuration from a blob object did not compile on
242 platforms with fgetc() etc. implemented as macros.
243 (merge 49d6cfa hv/config-from-blob later to maint-1.8.3).
245 * The recent "short-cut clone connectivity check" topic broke a
246 shallow repository when a fetch operation tries to auto-follow tags.
247 (merge 6da8bdc nd/fetch-pack-shallow-fix later to maint-1.8.3).