1 Git v2.15.1 Release Notes
2 =========================
7 * TravisCI build updates.
9 * "auto" as a value for the columnar output configuration ought to
10 judge "is the output consumed by humans?" with the same criteria as
11 "auto" for coloured output configuration, i.e. either the standard
12 output stream is going to tty, or a pager is in use. We forgot the
13 latter, which has been fixed.
15 * The experimental "color moved lines differently in diff output"
16 feature was buggy around "ignore whitespace changes" edges, which
19 * Instead of using custom line comparison and hashing functions to
20 implement "moved lines" coloring in the diff output, use the pair
21 of these functions from lower-layer xdiff/ code.
23 * Some codepaths did not check for errors when asking what branch the
24 HEAD points at, which have been fixed.
26 * "git commit", after making a commit, did not check for errors when
27 asking on what branch it made the commit, which has been corrected.
29 * "git status --ignored -u" did not stop at a working tree of a
30 separate project that is embedded in an ignored directory and
31 listed files in that other project, instead of just showing the
32 directory itself as ignored.
34 * A broken access to object databases in recent update to "git grep
35 --recurse-submodules" has been fixed.
37 * A recent regression in "git rebase -i" that broke execution of git
38 commands from subdirectories via "exec" instruction has been fixed.
40 * "git check-ref-format --branch @{-1}" bit a "BUG()" when run
41 outside a repository for obvious reasons; clarify the documentation
42 and make sure we do not even try to expand the at-mark magic in
43 such a case, but still call the validation logic for branch names.
45 * Command line completion (in contrib/) update.
47 * Description of blame.{showroot,blankboundary,showemail,date}
48 configuration variables have been added to "git config --help".
50 * After an error from lstat(), diff_populate_filespec() function
51 sometimes still went ahead and used invalid data in struct stat,
54 * UNC paths are also relevant in Cygwin builds and they are now
55 tested just like Mingw builds.
57 * Correct start-up sequence so that a repository could be placed
58 immediately under the root directory again (which was broken at
61 * The credential helper for libsecret (in contrib/) has been improved
62 to allow possibly prompting the end user to unlock secrets that are
63 currently locked (otherwise the secrets may not be loaded).
65 * Updates from GfW project.
67 * "git rebase -i" recently started misbehaving when a submodule that
68 is configured with 'submodule.<name>.ignore' is dirty; this has
71 * Some error messages did not quote filenames shown in it, which have
74 * Building with NO_LIBPCRE1_JIT did not disable it, which has been fixed.
76 * We used to add an empty alternate object database to the system
77 that does not help anything; it has been corrected.
79 * Error checking in "git imap-send" for empty response has been
82 * An ancient bug in "git apply --ignore-space-change" codepath has
85 * There was a recent semantic mismerge in the codepath to write out a
86 section of a configuration section, which has been corrected.
88 Also contains various documentation updates and code clean-ups.