1 Git v2.12.2 Release Notes
2 =========================
7 * "git status --porcelain" is supposed to give a stable output, but a
8 few strings were left as translatable by mistake.
10 * "Dumb http" transport used to misparse a nonsense http-alternates
11 response, which has been fixed.
13 * "git diff --quiet" relies on the size field in diff_filespec to be
14 correctly populated, but diff_populate_filespec() helper function
15 made an incorrect short-cut when asked only to populate the size
16 field for paths that need to go through convert_to_git() (e.g. CRLF
19 * There is no need for Python only to give a few messages to the
20 standard error stream, but we somehow did.
22 * A leak in a codepath to read from a packed object in (rare) cases
25 * "git upload-pack", which is a counter-part of "git fetch", did not
26 report a request for a ref that was not advertised as invalid.
27 This is generally not a problem (because "git fetch" will stop
28 before making such a request), but is the right thing to do.
30 * A "gc.log" file left by a backgrounded "gc --auto" disables further
31 automatic gc; it has been taught to run at least once a day (by
32 default) by ignoring a stale "gc.log" file that is too old.
34 * "git remote rm X", when a branch has remote X configured as the
35 value of its branch.*.remote, tried to remove branch.*.remote and
36 branch.*.merge and failed if either is unset.
38 * A caller of tempfile API that uses stdio interface to write to
39 files may ignore errors while writing, which is detected when
40 tempfile is closed (with a call to ferror()). By that time, the
41 original errno that may have told us what went wrong is likely to
42 be long gone and was overwritten by an irrelevant value.
43 close_tempfile() now resets errno to EIO to make errno at least
46 * "git show-branch" expected there were only very short branch names
47 in the repository and used a fixed-length buffer to hold them
48 without checking for overflow.
50 * The code that parses header fields in the commit object has been
51 updated for (micro)performance and code hygiene.
53 * A test that creates a confusing branch whose name is HEAD has been
54 corrected not to do so.
56 * "Cc:" on the trailer part does not have to conform to RFC strictly,
57 unlike in the e-mail header. "git send-email" has been updated to
58 ignore anything after '>' when picking addresses, to allow non-address
59 cruft like " # stable 4.4" after the address.
61 * "git push" had a handful of codepaths that could lead to a deadlock
62 when unexpected error happened, which has been fixed.
64 * Code to read submodule.<name>.ignore config did not state the
65 variable name correctly when giving an error message diagnosing
68 * "git ls-remote" and "git archive --remote" are designed to work
69 without being in a directory under Git's control. However, recent
70 updates revealed that we randomly look into a directory called
71 .git/ without actually doing necessary set-up when working in a
72 repository. Stop doing so.
74 * The code to parse the command line "git grep <patterns>... <rev>
75 [[--] <pathspec>...]" has been cleaned up, and a handful of bugs
76 have been fixed (e.g. we used to check "--" if it is a rev).
78 * The code to parse "git -c VAR=VAL cmd" and set configuration
79 variable for the duration of cmd had two small bugs, which have
81 This supersedes jc/config-case-cmdline topic that has been discarded.
83 Also contains various documentation updates and code clean-ups.