| 1 | Git v2.8.3 Release Notes |
| 2 | ======================== |
| 3 | |
| 4 | Fixes since v2.8.2 |
| 5 | ------------------ |
| 6 | |
| 7 | * "git send-email" now uses a more readable timestamps when |
| 8 | formulating a message ID. |
| 9 | |
| 10 | * The repository set-up sequence has been streamlined (the biggest |
| 11 | change is that there is no longer git_config_early()), so that we |
| 12 | do not attempt to look into refs/* when we know we do not have a |
| 13 | Git repository. |
| 14 | |
| 15 | * When "git worktree" feature is in use, "git branch -d" allowed |
| 16 | deletion of a branch that is checked out in another worktree |
| 17 | |
| 18 | * When "git worktree" feature is in use, "git branch -m" renamed a |
| 19 | branch that is checked out in another worktree without adjusting |
| 20 | the HEAD symbolic ref for the worktree. |
| 21 | |
| 22 | * "git format-patch --help" showed `-s` and `--no-patch` as if these |
| 23 | are valid options to the command. We already hide `--patch` option |
| 24 | from the documentation, because format-patch is about showing the |
| 25 | diff, and the documentation now hides these options as well. |
| 26 | |
| 27 | * A change back in version 2.7 to "git branch" broke display of a |
| 28 | symbolic ref in a non-standard place in the refs/ hierarchy (we |
| 29 | expect symbolic refs to appear in refs/remotes/*/HEAD to point at |
| 30 | the primary branch the remote has, and as .git/HEAD to point at the |
| 31 | branch we locally checked out). |
| 32 | |
| 33 | * A partial rewrite of "git submodule" in the 2.7 timeframe changed |
| 34 | the way the gitdir: pointer in the submodules point at the real |
| 35 | repository location to use absolute paths by accident. This has |
| 36 | been corrected. |
| 37 | |
| 38 | * "git commit" misbehaved in a few minor ways when an empty message |
| 39 | is given via -m '', all of which has been corrected. |
| 40 | |
| 41 | * Support for CRAM-MD5 authentication method in "git imap-send" did |
| 42 | not work well. |
| 43 | |
| 44 | * The socks5:// proxy support added back in 2.6.4 days was not aware |
| 45 | that socks5h:// proxies behave differently. |
| 46 | |
| 47 | * "git config" had a codepath that tried to pass a NULL to |
| 48 | printf("%s"), which nobody seems to have noticed. |
| 49 | |
| 50 | * On Cygwin, object creation uses the "create a temporary and then |
| 51 | rename it to the final name" pattern, not "create a temporary, |
| 52 | hardlink it to the final name and then unlink the temporary" |
| 53 | pattern. |
| 54 | |
| 55 | This is necessary to use Git on Windows shared directories, and is |
| 56 | already enabled for the MinGW and plain Windows builds. It also |
| 57 | has been used in Cygwin packaged versions of Git for quite a while. |
| 58 | See https://lore.kernel.org/git/20160419091055.GF2345@dinwoodie.org/ |
| 59 | and https://lore.kernel.org/git/20150811100527.GW14466@dinwoodie.org/. |
| 60 | |
| 61 | * "git replace -e" did not honour "core.editor" configuration. |
| 62 | |
| 63 | * Upcoming OpenSSL 1.1.0 will break compilation b updating a few APIs |
| 64 | we use in imap-send, which has been adjusted for the change. |
| 65 | |
| 66 | * "git submodule" reports the paths of submodules the command |
| 67 | recurses into, but this was incorrect when the command was not run |
| 68 | from the root level of the superproject. |
| 69 | |
| 70 | * The test scripts for "git p4" (but not "git p4" implementation |
| 71 | itself) has been updated so that they would work even on a system |
| 72 | where the installed version of Python is python 3. |
| 73 | |
| 74 | * The "user.useConfigOnly" configuration variable makes it an error |
| 75 | if users do not explicitly set user.name and user.email. However, |
| 76 | its check was not done early enough and allowed another error to |
| 77 | trigger, reporting that the default value we guessed from the |
| 78 | system setting was unusable. This was a suboptimal end-user |
| 79 | experience as we want the users to set user.name/user.email without |
| 80 | relying on the auto-detection at all. |
| 81 | |
| 82 | * "git mv old new" did not adjust the path for a submodule that lives |
| 83 | as a subdirectory inside old/ directory correctly. |
| 84 | |
| 85 | * "git push" from a corrupt repository that attempts to push a large |
| 86 | number of refs deadlocked; the thread to relay rejection notices |
| 87 | for these ref updates blocked on writing them to the main thread, |
| 88 | after the main thread at the receiving end notices that the push |
| 89 | failed and decides not to read these notices and return a failure. |
| 90 | |
| 91 | * A question by "git send-email" to ask the identity of the sender |
| 92 | has been updated. |
| 93 | |
| 94 | * Recent update to Git LFS broke "git p4" by changing the output from |
| 95 | its "lfs pointer" subcommand. |
| 96 | |
| 97 | * Some multi-byte encoding can have a backslash byte as a later part |
| 98 | of one letter, which would confuse "highlight" filter used in |
| 99 | gitweb. |
| 100 | |
| 101 | Also contains minor documentation updates and code clean-ups. |