| 1 | Git v2.6.5 Release Notes |
| 2 | ======================== |
| 3 | |
| 4 | Fixes since v2.6.4 |
| 5 | ------------------ |
| 6 | |
| 7 | * Because "test_when_finished" in our test framework queues the |
| 8 | clean-up tasks to be done in a shell variable, it should not be |
| 9 | used inside a subshell. Add a mechanism to allow 'bash' to catch |
| 10 | such uses, and fix the ones that were found. |
| 11 | |
| 12 | * Update "git subtree" (in contrib/) so that it can take whitespaces |
| 13 | in the pathnames, not only in the in-tree pathname but the name of |
| 14 | the directory that the repository is in. |
| 15 | |
| 16 | * Cosmetic improvement to lock-file error messages. |
| 17 | |
| 18 | * mark_tree_uninteresting() has code to handle the case where it gets |
| 19 | passed a NULL pointer in its 'tree' parameter, but the function had |
| 20 | 'object = &tree->object' assignment before checking if tree is |
| 21 | NULL. This gives a compiler an excuse to declare that tree will |
| 22 | never be NULL and apply a wrong optimization. Avoid it. |
| 23 | |
| 24 | * The helper used to iterate over loose object directories to prune |
| 25 | stale objects did not closedir() immediately when it is done with a |
| 26 | directory--a callback such as the one used for "git prune" may want |
| 27 | to do rmdir(), but it would fail on open directory on platforms |
| 28 | such as WinXP. |
| 29 | |
| 30 | * "git p4" used to import Perforce CLs that touch only paths outside |
| 31 | the client spec as empty commits. It has been corrected to ignore |
| 32 | them instead, with a new configuration git-p4.keepEmptyCommits as a |
| 33 | backward compatibility knob. |
| 34 | |
| 35 | * The exit code of git-fsck did not reflect some types of errors |
| 36 | found in packed objects, which has been corrected. |
| 37 | |
| 38 | * The completion script (in contrib/) used to list "git column" |
| 39 | (which is not an end-user facing command) as one of the choices |
| 40 | |
| 41 | * Improve error reporting when SMTP TLS fails. |
| 42 | |
| 43 | * When getpwuid() on the system returned NULL (e.g. the user is not |
| 44 | in the /etc/passwd file or other uid-to-name mappings), the |
| 45 | codepath to find who the user is to record it in the reflog barfed |
| 46 | and died. Loosen the check in this codepath, which already accepts |
| 47 | questionable ident string (e.g. host part of the e-mail address is |
| 48 | obviously bogus), and in general when we operate fmt_ident() function |
| 49 | in non-strict mode. |
| 50 | |
| 51 | * "git symbolic-ref" forgot to report a failure with its exit status. |
| 52 | |
| 53 | * History traversal with "git log --source" that starts with an |
| 54 | annotated tag failed to report the tag as "source", due to an |
| 55 | old regression in the command line parser back in v2.2 days. |
| 56 | |
| 57 | Also contains typofixes, documentation updates and trivial code |
| 58 | clean-ups. |