1Git v1.8.5 Release Notes 2======================== 3 4Backward compatibility notes (for Git 2.0) 5------------------------------------------ 6 7When "git push [$there]" does not say what to push, we have used the 8traditional "matching" semantics so far (all your branches were sent 9to the remote as long as there already are branches of the same name 10over there). In Git 2.0, the default will change to the "simple" 11semantics that pushes: 12 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 16 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. 19 20Use the user preference configuration variable "push.default" to 21change this. If you are an old-timer who is used to the "matching" 22semantics, you can set the variable to "matching" to keep the 23traditional behaviour. If you want to live in the future early, you 24can set it to "simple" today without waiting for Git 2.0. 25 26When "git add -u" (and "git add -A") is run inside a subdirectory and 27does not specify which paths to add on the command line, it 28will operate on the entire tree in Git 2.0 for consistency 29with "git commit -a" and other commands. There will be no 30mechanism to make plain "git add -u" behave like "git add -u .". 31Current users of "git add -u" (without a pathspec) should start 32training their fingers to explicitly say "git add -u ." 33before Git 2.0 comes. A warning is issued when these commands are 34run without a pathspec and when you have local changes outside the 35current directory, because the behaviour in Git 2.0 will be different 36from today's version in such a situation. 37 38In Git 2.0, "git add <path>" will behave as "git add -A <path>", so 39that "git add dir/" will notice paths you removed from the directory 40and record the removal. Versions before Git 2.0, including this 41release, will keep ignoring removals, but the users who rely on this 42behaviour are encouraged to start using "git add --ignore-removal <path>" 43now before 2.0 is released. 44 45 46Updates since v1.8.4 47-------------------- 48 49Foreign interfaces, subsystems and ports. 50 51 * remote-hg remote helper misbehaved when interacting with a local Hg 52 repository relative to the home directory, e.g. "clone hg::~/there". 53 54 * imap-send ported to OS X uses Apple's security framework instead of 55 OpenSSL one. 56 57 * Subversion 1.8.0 that was recently released breaks older subversion 58 clients coming over http/https in various ways. 59 60 * "git fast-import" treats an empty path given to "ls" as the root of 61 the tree. 62 63 64UI, Workflows & Features 65 66 * "git config" did not provide a way to set or access numbers larger 67 than a native "int" on the platform; it now provides 64-bit signed 68 integers on all platforms. 69 70 * "git pull --rebase" always chose to do the bog-standard flattening 71 rebase. You can tell it to run "rebase --preserve-merges" by 72 setting "pull.rebase" configuration to "preserve". 73 74 * "git push --no-thin" actually disables the "thin pack transfer" 75 optimization. 76 77 * Magic pathspecs like ":(icase)makefile" that matches both 78 Makefile and makefile can be used in more places. 79 80 * The "http.*" variables can now be specified per URL that the 81 configuration applies. For example, 82 83 [http] 84 sslVerify = true 85 [http "https://weak.example.com/"] 86 sslVerify = false 87 88 would flip http.sslVerify off only when talking to that specified 89 site. 90 91 * "git mv A B" when moving a submodule A has been taught to 92 relocate its working tree and to adjust the paths in the 93 .gitmodules file. 94 95 * "git blame" can now take more than one -L option to discover the 96 origin of multiple blocks of the lines. 97 98 * The http transport clients can optionally ask to save cookies 99 with http.savecookies configuration variable. 100 101 * "git push" learned a more fine grained control over a blunt 102 "--force" when requesting a non-fast-forward update with the 103 "--force-with-lease=<refname>:<expected object name>" option. 104 105 * "git diff --diff-filter=<classes of changes>" can now take 106 lowercase letters (e.g. "--diff-filter=d") to mean "show 107 everything but these classes". "git diff-files -q" is now a 108 deprecated synonym for "git diff-files --diff-filter=d". 109 110 * "git fetch" (hence "git pull" as well) learned to check 111 "fetch.prune" and "remote.*.prune" configuration variables and 112 to behave as if the "--prune" command line option was given. 113 114 * "git check-ignore -z" applied the NUL termination to both its input 115 (with --stdin) and its output, but "git check-attr -z" ignored the 116 option on the output side. Make both honor -z on the input and 117 output side the same way. 118 119 * "git whatchanged" may still be used by old timers, but mention of 120 it in documents meant for new users will only waste readers' time 121 wonderig what the difference is between it and "git log". Make it 122 less prominent in the general part of the documentation and explain 123 that it is merely a "git log" with different default behaviour in 124 its own document. 125 126 127Performance, Internal Implementation, etc. 128 129 * Many commands use --dashed-option as a operation mode selector 130 (e.g. "git tag --delete") that the user can use at most one 131 (e.g. "git tag --delete --verify" is a nonsense) and you cannot 132 negate (e.g. "git tag --no-delete" is a nonsense). parse-options 133 API learned a new OPT_CMDMODE macro to make it easier to implement 134 such a set of options. 135 136 * OPT_BOOLEAN() in parse-options API was misdesigned to be "counting 137 up" but many subcommands expect it to behave as "on/off". Update 138 them to use OPT_BOOL() which is a proper boolean. 139 140 * "git gc" exits early without doing a double-work when it detects 141 that another instance of itself is already running. 142 143 * Under memory pressure and/or file descriptor pressure, we used to 144 close pack windows that are not used and also closed filehandle to 145 an open but unused packfiles. These are now controlled separately 146 to better cope with the load. 147 148Also contains various documentation updates and code clean-ups. 149 150 151Fixes since v1.8.4 152------------------ 153 154Unless otherwise noted, all the fixes since v1.8.4 in the maintenance 155track are contained in this release (see release notes to them for 156details). 157 158 * "git ls-files -k" needs to crawl only the part of the working tree 159 that may overlap the paths in the index to find killed files, but 160 shared code with the logic to find all the untracked files, which 161 made it unnecessarily inefficient. 162 (merge 680be04 jc/ls-files-killed-optim later to maint). 163 164 * The commit object names in the insn sheet that was prepared at the 165 beginning of "rebase -i" session can become ambiguous as the 166 rebasing progresses and the repository gains more commits. Make 167 sure the internal record is kept with full 40-hex object names. 168 (merge 75c6976 es/rebase-i-no-abbrev later to maint). 169 170 * "git rebase --preserve-merges" internally used the merge machinery 171 and as a side effect, left merge summary message in the log, but 172 when rebasing, there should not be a need for merge summary. 173 (merge a9f739c rt/rebase-p-no-merge-summary later to maint). 174 175 * A call to xread() was used without a loop around to cope with short 176 read in the codepath to stream new contents to a pack. 177 (merge e92527c js/xread-in-full later to maint). 178 179 * "git rebase -i" forgot that the comment character can be 180 configurable while reading its insn sheet. 181 (merge 7bca7af es/rebase-i-respect-core-commentchar later to maint). 182 183 * The mailmap support code read past the allocated buffer when the 184 mailmap file ended with an incomplete line. 185 (merge f972a16 jk/mailmap-incomplete-line later to maint). 186 187 * We used to send a large request to read(2)/write(2) as a single 188 system call, which was bad from the latency point of view when 189 the operation needs to be killed, and also triggered an error on 190 broken 64-bit systems that refuse to take more than 2GB read or 191 write in one go. 192 (merge a487916 sp/clip-read-write-to-8mb later to maint). 193 194 * "git fetch" that auto-followed tags incorrectly reused the 195 connection with Git-aware transport helper (like the sample "ext::" 196 helper shipped with Git). 197 (merge 0f73f8b jc/transport-do-not-use-connect-twice-in-fetch later to maint). 198 199 * "git log --full-diff -- <pathspec>" showed a huge diff for paths 200 outside the given <pathspec> for each commit, instead of showing 201 the change relative to the parent of the commit. "git reflog -p" 202 had a similar problem. 203 (merge 838f9a1 tr/log-full-diff-keep-true-parents later to maint). 204 205 * Setting submodule.*.path configuration variable to true (without 206 giving "= value") caused Git to segfault. 207 (merge 4b05440 jl/some-submodule-config-are-not-boolean later to maint). 208 209 * "git rebase -i" (there could be others, as the root cause is pretty 210 generic) fed a random, data dependeant string to 'echo' and 211 expects it to come out literally, corrupting its error message. 212 (merge 89b0230 mm/no-shell-escape-in-die-message later to maint). 213 214 * Some people still use rather old versions of bash, which cannot 215 grok some constructs like 'printf -v varname' the prompt and 216 completion code started to use recently. 217 (merge a44aa69 bc/completion-for-bash-3.0 later to maint). 218 219 * Code to read configuration from a blob object did not compile on 220 platforms with fgetc() etc. implemented as macros. 221 (merge 49d6cfa hv/config-from-blob later to maint-1.8.3). 222 223 * The recent "short-cut clone connectivity check" topic broke a 224 shallow repository when a fetch operation tries to auto-follow tags. 225 (merge 6da8bdc nd/fetch-pack-shallow-fix later to maint-1.8.3).