1Git v1.8.0 Release Notes 2======================== 3 4Backward compatibility notes 5---------------------------- 6 7In the next major release, we will change the behaviour of the "git 8push" command. When "git push [$there]" does not say what to push, we 9have used the traditional "matching" semantics (all your branches were 10sent to the remote as long as there already are branches of the same 11name over there). We will 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 19Updates since v1.7.12 20--------------------- 21 22UI, Workflows & Features 23 24 * A credential helper for Win32 to allow access to the keychain of 25 the logged-in user has been added. 26 27 * "git cherry-pick" learned the "--allow-empty-message" option to 28 allow it to replay a commit without any log message. 29 30 * "git daemon" learned the "--access-hook" option to allow an 31 external command to decline service based on the client address, 32 repository path, etc. 33 34 * "git difftool --dir-diff" learned to use symbolic links to prepare 35 temporary copy of the working tree when available. 36 37 * "git grep" learned to use a non-standard pattern type by default if 38 a configuration variable tells it to. 39 40Foreign Interface 41 42 * "git svn" has been updated to work with SVN 1.7. 43 44Performance, Internal Implementation, etc. (please report possible regressions) 45 46 * The "check-docs" build target has been updated and greatly 47 simplified. 48 49 * The documentation in the TeXinfo format was using indented output 50 for materials meant to be examples that are better typeset in 51 monospace. 52 53 * Compatibility wrapper to learn the maximum number of file 54 descriptors we can open around sysconf(_SC_OPEN_MAX) and 55 getrlimit(RLIMIT_NO_FILE) has been introduced for portability. 56 57 * Compatibility wrapper around some mkdir(2) implementations that 58 reject parameter with trailing slash has been introduced. 59 60 * The heuristics to detect and silently convert latin1 to utf8 when 61 we were told to use utf-8 in the log message has been transplanted 62 from "mailinfo" to "commit" and "commit-tree". 63 64 * Messages given by "git <subcommand> -h" from many subcommands have 65 been marked for translation. 66 67Also contains minor documentation updates and code clean-ups. 68 69 70Fixes since v1.7.12 71------------------- 72 73Unless otherwise noted, all the fixes since v1.7.12 in the 74maintenance track are contained in this release (see release notes 75to them for details). 76 77 * "git p4", when "--use-client-spec" and "--detect-branches" are used 78 together, misdetected branches. 79 (merge 21ef5df pw/p4-use-client-spec-branch-detection later to maint). 80 81 * Output from "git branch -v" contains "(no branch)" that could be 82 localized, but the code to align it along with the names of 83 branches were counting in bytes, not in display columns. 84 (merge 1452bd6 nd/branch-v-alignment later to maint). 85 86 * "git for-each-ref" did not currectly support more than one --sort 87 option. 88 (merge 3b51222 kk/maint-for-each-ref-multi-sort later to maint). 89 90 * Pushing to smart HTTP server with recent Git fails without having 91 the username in the URL to force authentication, if the server is 92 configured to allow GET anonymously, while requiring authentication 93 for POST. 94 (merge b81401c jk/maint-http-half-auth-push later to maint). 95 96 * When looking for $HOME/.gitconfig etc., it is OK if we cannot read 97 them because they do not exist, but we did not diagnose existing 98 files that we cannot read. 99 100 * The synopsis said "checkout [-B branch]" to make it clear the 101 branch name is a parameter to the option, but the heading for the 102 option description was "-B::", not "-B branch::", making the 103 documentation misleading. There may be room in documentation pages 104 of other commands for similar improvements (hint, hint). 105 (merge 45aaf03 jc/maint-doc-checkout-b-always-takes-branch-name later to maint). 106 107 * "git log .." errored out saying it is both rev range and a path 108 when there is no disambiguating "--" is on the command line. 109 Update the command line parser to interpret ".." as a path in such 110 a case. 111 (merge 003c84f jc/dotdot-is-parent-directory later to maint). 112 113 * "git apply -p0" did not parse pathnames on "diff --git" line 114 correctly. This caused patches that had pathnames in no other 115 places to be mistakenly rejected (most notably, binary patch that 116 does not rename nor change mode). Textual patches, renames or mode 117 changes have preimage and postimage pathnames in different places 118 in a form that can be parsed unambiguously and did not suffer from 119 this problem. 120 (merge 6a2abdc jc/apply-binary-p0 later to maint). 121 122 * The exit status code from "git config" was way overspecified while 123 being incorrect. The implementation has been updated to give the 124 documented status for a case that was documented, and introduce a 125 new code for "all other errors". 126 (merge 9409c7a jc/maint-config-exit-status later to maint). 127 128 * "git foo" errored out with "Not a directory" when the user had a 129 non-directory on $PATH, and worse yet it masked an alias "foo" from 130 running. (merge a785508 jc/maint-sane-execvp-notdir later to 131 maint). 132 133 * The interactive prompt "git send-email" gives was error prone. It 134 asked "What e-mail address do you want to use?" with the address it 135 guessed (correctly) the user would want to use in its prompt, 136 tempting the user to say "y". But the response was taken as "No, 137 please use 'y' as the e-mail address instead", which is most 138 certainly not what the user meant. 139 (merge 51bbccf jc/send-email-reconfirm later to maint). 140 141 * "git stash apply/pop" did not trigger "rerere" upon conflicts 142 unlike other mergy operations. 143 (merge 743bf6d ph/stash-rerere later to maint). 144 145 * "git submodule update -f" did not update paths in the working tree 146 that has local changes. 147 (merge 01d4721 sz/submodule-force-update later to maint). 148 149 * We used curl_easy_strerror() without checking version of cURL, 150 breaking the build for versions before curl 7.12.0. 151 (merge 4246b0b js/no-curl-easy-strerror-on-old-curl later to maint). 152 153 * Code to work around MacOS X UTF-8 gotcha has been cleaned up. 154 (merge 9a27f96 rr/precompose-utf8-cleanup later to maint). 155 156 * Documentation for the configuration file format had a confusing 157 example. 158 (merge d1e1fe7 mh/maint-config-doc-proxy-command later to maint). 159 160 * "git submodule <cmd> path" did not error out when the path to the 161 submodule was misspelt. 162 (merge be9d0a3 hv/submodule-path-unmatch later to maint). 163 164 * Some capabilities were asked by fetch-pack even when upload-pack 165 did not advertise that they are available. fetch-pack has been 166 fixed not to do so. 167 168 * The output from "git diff -B" for a file that ends with an 169 incomplete line did not put "\ No newline..." on a line of its own.