1Git v1.8.1 Release Notes 2======================== 3 4Backward compatibility notes 5---------------------------- 6 7In the next major release (not *this* one), we will change the 8behavior of the "git push" command. 9 10When "git push [$there]" does not say what to push, we have used the 11traditional "matching" semantics so far (all your branches were sent 12to the remote as long as there already are branches of the same name 13over there). We will use the "simple" semantics that pushes the 14current branch to the branch with the same name, only when the current 15branch is set to integrate with that remote branch. There is a user 16preference configuration variable "push.default" to change this, and 17"git push" will warn about the upcoming change until you set this 18variable in this release. 19 20"git branch --set-upstream" is deprecated and may be removed in a 21relatively distant future. "git branch [-u|--set-upstream-to]" has 22been introduced with a saner order of arguments to replace it. 23 24 25Updates since v1.8.0 26-------------------- 27 28UI, Workflows & Features 29 30 * Command-line completion scripts for tcsh and zsh have been added. 31 32 * A new remote-helper interface for Mercurial has been added to 33 contrib/remote-helpers. 34 35 * We used to have a workaround for a bug in ancient "less" that 36 causes it to exit without any output when the terminal is resized. 37 The bug has been fixed in "less" version 406 (June 2007), and the 38 workaround has been removed in this release. 39 40 * Some documentation pages that used to ship only in the plain text 41 format are now formatted in HTML as well. 42 43 * "git-prompt" scriptlet (in contrib/completion) can be told to paint 44 pieces of the hints in the prompt string in colors. 45 46 * A new configuration variable "diff.context" can be used to 47 give the default number of context lines in the patch output, to 48 override the hardcoded default of 3 lines. 49 50 * When "git checkout" checks out a branch, it tells the user how far 51 behind (or ahead) the new branch is relative to the remote tracking 52 branch it builds upon. The message now also advises how to sync 53 them up by pushing or pulling. This can be disabled with the 54 advice.statusHints configuration variable. 55 56 * "git config --get" used to diagnose presence of multiple 57 definitions of the same variable in the same configuration file as 58 an error, but it now applies the "last one wins" rule used by the 59 internal configuration logic. Strictly speaking, this may be an 60 API regression but it is expected that nobody will notice it in 61 practice. 62 63 * "git log -p -S<string>" now looks for the <string> after applying 64 the textconv filter (if defined); earlier it inspected the contents 65 of the blobs without filtering. 66 67 * "git format-patch" learned the "--notes=<ref>" option to give 68 notes for the commit after the three-dash lines in its output. 69 70 * "git log --grep=<pcre>" learned to honor the "grep.patterntype" 71 configuration set to "perl". 72 73 * "git replace -d <object>" now interprets <object> as an extended 74 SHA-1 (e.g. HEAD~4 is allowed), instead of only accepting full hex 75 object name. 76 77 * "git rm $submodule" used to punt on removing a submodule working 78 tree to avoid losing the repository embedded in it. Because 79 recent git uses a mechanism to separate the submodule repository 80 from the submodule working tree, "git rm" learned to detect this 81 case and removes the submodule working tree when it is safe to do so. 82 83 * "git send-email" used to prompt for the sender address, even when 84 the committer identity is well specified (e.g. via user.name and 85 user.email configuration variables). The command no longer gives 86 this prompt when not necessary. 87 88 * "git send-email" did not allow non-address garbage strings to 89 appear after addresses on Cc: lines in the patch files (and when 90 told to pick them up to find more recipients), e.g. 91 92 Cc: Stable Kernel <stable@k.org> # for v3.2 and up 93 94 The command now strips " # for v3.2 and up" part before adding the 95 remainder of this line to the list of recipients. 96 97 * "git submodule add" learned to add a new submodule at the same 98 path as the path where an unrelated submodule was bound to in an 99 existing revision via the "--name" option. 100 101 * "git submodule sync" learned the "--recursive" option. 102 103 * "diff.submodule" configuration variable can be used to give custom 104 default value to the "git diff --submodule" option. 105 106 * "git symbolic-ref" learned the "-d $symref" option to delete the 107 named symbolic ref, which is more intuitive way to spell it than 108 "update-ref -d --no-deref $symref". 109 110 111Foreign Interface 112 113 * "git cvsimport" can be told to record timezones (other than GMT) 114 per-author via its author info file. 115 116 * The remote helper interface to interact with subversion 117 repositories (one of the GSoC 2012 projects) has been merged. 118 119 120Performance, Internal Implementation, etc. 121 122 * Compilation on Cygwin with newer header files are supported now. 123 124 * The logic to generate the initial advertisement from "upload-pack" 125 (i.e. what is invoked by "git fetch" on the other side of the 126 connection) to list what refs are available in the repository has 127 been optimized. 128 129 * The logic to find set of attributes that match a given path has 130 been optimized. 131 132 * Use preloadindex in "git diff-index" and "git update-index", which 133 has a nice speedup on systems with slow stat calls (and even on 134 Linux). 135 136 137Also contains minor documentation updates and code clean-ups. 138 139 140Fixes since v1.8.0 141------------------ 142 143Unless otherwise noted, all the fixes since v1.8.0 in the maintenance 144track are contained in this release (see release notes to them for 145details). 146 147 * The configuration parser had an unnecessary hardcoded limit on 148 variable names that was not checked consistently. 149 150 * The "say" function in the test scaffolding incorrectly allowed 151 "echo" to interpret "\a" as if it were a C-string asking for a 152 BEL output. 153 154 * "git mergetool" feeds /dev/null as a common ancestor when dealing 155 with an add/add conflict, but p4merge backend cannot handle 156 it. Work it around by passing a temporary empty file. 157 158 * "git log -F -E --grep='<ere>'" failed to use the given <ere> 159 pattern as extended regular expression, and instead looked for the 160 string literally. 161 162 * "git grep -e pattern <tree>" asked the attribute system to read 163 "<tree>:.gitattributes" file in the working tree, which was 164 nonsense. 165 166 * A symbolic ref refs/heads/SYM was not correctly removed with "git 167 branch -d SYM"; the command removed the ref pointed by SYM 168 instead. 169 170 * Update "remote tracking branch" in the documentation to 171 "remote-tracking branch". 172 173 * "git pull --rebase" run while the HEAD is detached tried to find 174 the upstream branch of the detached HEAD (which by definition 175 does not exist) and emitted unnecessary error messages. 176 177 * The refs/replace hierarchy was not mentioned in the 178 repository-layout docs. 179 180 * Various rfc2047 quoting issues around a non-ASCII name on the 181 From: line in the output from format-patch have been corrected. 182 183 * Sometimes curl_multi_timeout() function suggested a wrong timeout 184 value when there is no file descriptor to wait on and the http 185 transport ended up sleeping for minutes in select(2) system call. 186 A workaround has been added for this. 187 188 * For a fetch refspec (or the result of applying wildcard on one), 189 we always want the RHS to map to something inside "refs/" 190 hierarchy, but the logic to check it was not exactly right. 191 (merge 5c08c1f jc/maint-fetch-tighten-refname-check later to maint). 192 193 * "git diff -G<pattern>" did not honor textconv filter when looking 194 for changes. 195 196 * Some HTTP servers ask for auth only during the actual packing phase 197 (not in ls-remote phase); this is not really a recommended 198 configuration, but the clients used to fail to authenticate with 199 such servers. 200 (merge 2e736fd jk/maint-http-half-auth-fetch later to maint). 201 202 * "git p4" used to try expanding malformed "$keyword$" that spans 203 across multiple lines. 204 205 * Syntax highlighting in "gitweb" was not quite working. 206 207 * RSS feed from "gitweb" had a xss hole in its title output. 208 209 * "git config --path $key" segfaulted on "[section] key" (a boolean 210 "true" spelled without "=", not "[section] key = true"). 211 212 * "git checkout -b foo" while on an unborn branch did not say 213 "Switched to a new branch 'foo'" like other cases. 214 215 * Various codepaths have workaround for a common misconfiguration to 216 spell "UTF-8" as "utf8", but it was not used uniformly. Most 217 notably, mailinfo (which is used by "git am") lacked this support. 218 219 * We failed to mention a file without any content change but whose 220 permission bit was modified, or (worse yet) a new file without any 221 content in the "git diff --stat" output. 222 223 * When "--stat-count" hides a diffstat for binary contents, the total 224 number of added and removed lines at the bottom was computed 225 incorrectly. 226 227 * When "--stat-count" hides a diffstat for unmerged paths, the total 228 number of affected files at the bottom of the "diff --stat" output 229 was computed incorrectly. 230 231 * "diff --shortstat" miscounted the total number of affected files 232 when there were unmerged paths. 233 234 * "update-ref -d --deref SYM" to delete a ref through a symbolic ref 235 that points to it did not remove it correctly.