1 Git v1.8.2.1 Release Notes
2 ==========================
7 * An earlier change to the attribute system introduced at v1.8.1.2 by
8 mistake stopped a pattern "dir" (without trailing slash) from
9 matching a directory "dir" (it only wanted to allow pattern "dir/"
12 * Verification of signed tags were not done correctly when not in C
15 * 'git commit -m "$msg"' used to add an extra newline even when
16 $msg already ended with one.
18 * The "--match=<pattern>" option of "git describe", when used with
19 "--all" to allow refs that are not annotated tags to be used as a
20 base of description, did not restrict the output from the command
21 to those that match the given pattern.
23 * An aliased command spawned from a bare repository that does not say
24 it is bare with "core.bare = yes" is treated as non-bare by mistake.
26 * When "format-patch" quoted a non-ascii strings on the header files,
27 it incorrectly applied rfc2047 and chopped a single character in
30 * "git archive" reports a failure when asked to create an archive out
31 of an empty tree. It would be more intuitive to give an empty
32 archive back in such a case.
34 * "git tag -f <tag>" always said "Updated tag '<tag>'" even when
35 creating a new tag (i.e. not overwriting nor updating).
37 * "git cmd -- ':(top'" was not diagnosed as an invalid syntax, and
38 instead the parser kept reading beyond the end of the string.
40 * Annotated tags outside refs/tags/ hierarchy were not advertised
41 correctly to the ls-remote and fetch with recent version of Git.
43 * The code to keep track of what directory names are known to Git on
44 platforms with case insensitive filesystems can get confused upon a
45 hash collision between these pathnames and looped forever.
47 * The logic used by "git diff -M --stat" to shorten the names of
48 files before and after a rename did not work correctly when the
49 common prefix and suffix between the two filenames overlapped.
51 * "git submodule update", when recursed into sub-submodules, did not
52 accumulate the prefix paths.
54 * "git am $maildir/" applied messages in an unexpected order; sort
55 filenames read from the maildir/ in a way that is more likely to
56 sort messages in the order the writing MUA meant to, by sorting
57 numeric segment in numeric order and non-numeric segment in
60 * When export-subst is used, "zip" output recorded incorrect
63 * Some platforms and users spell UTF-8 differently; retry with the
64 most official "UTF-8" when the system does not understand the
65 user-supplied encoding name that are the common alternative
68 * "git branch" did not bother to check nonsense command line
69 parameters and issue errors in many cases.
71 * "git update-index -h" did not do the usual "-h(elp)" thing.
73 * perl/Git.pm::cat_blob slurped everything in core only to write it
74 out to a file descriptor, which was not a very smart thing to do.
76 * The SSL peer verification done by "git imap-send" did not ask for
77 Server Name Indication (RFC 4366), failing to connect SSL/TLS
78 sites that serve multiple hostnames on a single IP.
80 * "git index-pack" had a buffer-overflow while preparing an
81 informational message when the translated version of it was too
84 * Clarify in the documentation "what" gets pushed to "where" when the
85 command line to "git push" does not say these explicitly.
87 * In "git reflog expire", REACHABLE bit was not cleared from the
90 * The "--color=<when>" argument to the commands in the diff family
93 * The arguments given to pre-rebase hook were not documented.
95 * The v4 index format was not documented.
97 * The "--match=<pattern>" argument "git describe" takes uses glob
98 pattern but it wasn't obvious from the documentation.
100 * Some sources failed to compile on systems that lack NI_MAXHOST in
101 their system header (e.g. z/OS).
103 * Add an example use of "--env-filter" in "filter-branch"
106 * "git bundle verify" did not say "records a complete history" for a
107 bundle that does not have any prerequisites.
109 * In the v1.8.0 era, we changed symbols that do not have to be global
110 to file scope static, but a few functions in graph.c were used by
111 CGit from sideways bypassing the entry points of the API the
114 * "git merge-tree" had a typo in the logic to detect d/f conflicts,
115 which caused it to segfault in some cases.