| 1 | Git v2.4.3 Release Notes |
| 2 | ======================== |
| 3 | |
| 4 | Fixes since v2.4.3 |
| 5 | ------------------ |
| 6 | |
| 7 | * Error messages from "git branch" called remote-tracking branches as |
| 8 | "remote branches". |
| 9 | |
| 10 | * "git rerere forget" in a repository without rerere enabled gave a |
| 11 | cryptic error message; it should be a silent no-op instead. |
| 12 | |
| 13 | * "git pull --log" and "git pull --no-log" worked as expected, but |
| 14 | "git pull --log=20" did not. |
| 15 | |
| 16 | * The pull.ff configuration was supposed to override the merge.ff |
| 17 | configuration, but it didn't. |
| 18 | |
| 19 | * The code to read pack-bitmap wanted to allocate a few hundred |
| 20 | pointers to a structure, but by mistake allocated and leaked memory |
| 21 | enough to hold that many actual structures. Correct the allocation |
| 22 | size and also have it on stack, as it is small enough. |
| 23 | |
| 24 | * Various documentation mark-up fixes to make the output more |
| 25 | consistent in general and also make AsciiDoctor (an alternative |
| 26 | formatter) happier. |
| 27 | |
| 28 | * "git bundle verify" did not diagnose extra parameters on the |
| 29 | command line. |
| 30 | |
| 31 | * Multi-ref transaction support we merged a few releases ago |
| 32 | unnecessarily kept many file descriptors open, risking to fail with |
| 33 | resource exhaustion. |
| 34 | |
| 35 | * The ref API did not handle cases where 'refs/heads/xyzzy/frotz' is |
| 36 | removed at the same time as 'refs/heads/xyzzy' is added (or vice |
| 37 | versa) very well. |
| 38 | |
| 39 | * The "log --decorate" enhancement in Git 2.4 that shows the commit |
| 40 | at the tip of the current branch e.g. "HEAD -> master", did not |
| 41 | work with --decorate=full. |
| 42 | |
| 43 | * There was a commented-out (instead of being marked to expect |
| 44 | failure) test that documented a breakage that was fixed since the |
| 45 | test was written; turn it into a proper test. |
| 46 | |
| 47 | * core.excludesfile (defaulting to $XDG_HOME/git/ignore) is supposed |
| 48 | to be overridden by repository-specific .git/info/exclude file, but |
| 49 | the order was swapped from the beginning. This belatedly fixes it. |
| 50 | |
| 51 | * The connection initiation code for "ssh" transport tried to absorb |
| 52 | differences between the stock "ssh" and Putty-supplied "plink" and |
| 53 | its derivatives, but the logic to tell that we are using "plink" |
| 54 | variants were too loose and falsely triggered when "plink" appeared |
| 55 | anywhere in the path (e.g. "/home/me/bin/uplink/ssh"). |
| 56 | |
| 57 | * "git rebase -i" moved the "current" command from "todo" to "done" a |
| 58 | bit too prematurely, losing a step when a "pick" did not even start. |
| 59 | |
| 60 | * "git add -e" did not allow the user to abort the operation by |
| 61 | killing the editor. |
| 62 | |
| 63 | * Git 2.4 broke setting verbosity and progress levels on "git clone" |
| 64 | with native transports. |
| 65 | |
| 66 | * Some time ago, "git blame" (incorrectly) lost the convert_to_git() |
| 67 | call when synthesizing a fake "tip" commit that represents the |
| 68 | state in the working tree, which broke folks who record the history |
| 69 | with LF line ending to make their project portable across |
| 70 | platforms while terminating lines in their working tree files with |
| 71 | CRLF for their platform. |
| 72 | |
| 73 | * Code clean-up for xdg configuration path support. |
| 74 | |
| 75 | Also contains typofixes, documentation updates and trivial code |
| 76 | clean-ups. |