@@ -4,59 +4,81 @@ Git v1.7.11.6 Release Notes
4
4
Fixes since v1.7.11.5
5
5
---------------------
6
6
7
- This consists primarily of documentation updates and low-impact code
8
- clarification and bugfixes.
9
-
10
- - "ciabot" script (in contrib/) has been updated with extensive
7
+ * "ciabot" script (in contrib/) has been updated with extensive
11
8
documentation.
12
9
13
- - The "--rebase" option to "git pull" can be abbreviated to "-r",
14
- but we didn't document it.
10
+ * "git foo" errored out with "Not a directory" when the user had a
11
+ non-directory on $PATH, and worse yet it masked an alias "foo" from
12
+ running.
15
13
16
- - It was generally understood that "--long-option"s to many of our
17
- subcommands can be abbreviated to the unique prefix, but it was not
18
- easy to find it described for new readers of the documentation set .
14
+ * When the user exports a non-default IFS without HT, scripts that
15
+ rely on being able to parse "ls-files -s | while read a b c..."
16
+ started to fail. Protect them from such a misconfiguration .
19
17
20
- - The "--topo-order", "--date-order" (and the lack of either means
21
- the default order) options to "rev-list" and "log" family of
22
- commands were poorly described in the documentation.
18
+ * When the user gives an argument that can be taken as both a
19
+ revision name and a pathname without disambiguating with "--", we
20
+ used to give a help message "Use '--' to separate". The message
21
+ has been clarified to show where that '--' goes on the command
22
+ line.
23
+
24
+ * Documentation for the configuration file format had a confusing
25
+ example.
23
26
24
- - Older parts of the documentation described as if having a regular
27
+ * Older parts of the documentation described as if having a regular
25
28
file in .git/refs/ hierarchy were the only way to have branches and
26
29
tags, which is not true for quite some time.
27
30
28
- - A utility shell function test_seq has been added as a replacement
29
- for the 'seq' utility found on some platforms.
31
+ * It was generally understood that "--long-option"s to many of our
32
+ subcommands can be abbreviated to the unique prefix, but it was not
33
+ easy to find it described for new readers of the documentation set.
30
34
31
- - Fallback 'getpass' implementation made unportable use of stdio API.
35
+ * The "--topo-order", "--date-order" (and the lack of either means
36
+ the default order) options to "rev-list" and "log" family of
37
+ commands were poorly described in the documentation.
32
38
33
- - "git commit --amend" let the user edit the log message and then
39
+ * "git commit --amend" let the user edit the log message and then
34
40
died when the human-readable committer name was given
35
41
insufficiently by getpwent(3).
36
42
37
- - The reflog entries left by "git rebase" and "git rebase -i" were
38
- inconsistent (the interactive one gave an abbreviated object name).
43
+ * The exit status code from "git config" was way overspecified while
44
+ being incorrect. The implementation has been updated to give the
45
+ documented status for a case that was documented, and introduce a
46
+ new code for "all other errors".
39
47
40
- - When the user exports a non-default IFS without HT, scripts that
41
- rely on being able to parse "ls-files -s | while read a b c..."
42
- started to fail. Protect them from such a misconfiguration.
48
+ * The output from "git diff -B" for a file that ends with an
49
+ incomplete line did not put "\ No newline..." on a line of its own.
50
+
51
+ * "git diff" had a confusion between taking data from a path in the
52
+ working tree and taking data from an object that happens to have
53
+ name 0{40} recorded in a tree.
43
54
44
- - When "git push" triggered the automatic gc on the receiving end, a
55
+ * The "--rebase" option to "git pull" can be abbreviated to "-r",
56
+ but we didn't document it.
57
+
58
+ * When "git push" triggered the automatic gc on the receiving end, a
45
59
message from "git prune" that said it was removing cruft leaked to
46
60
the standard output, breaking the communication protocol.
47
61
48
- - "git diff" had a confusion between taking data from a path in the
49
- working tree and taking data from an object that happens to have
50
- name 0{40} recorded in a tree.
62
+ * The reflog entries left by "git rebase" and "git rebase -i" were
63
+ inconsistent (the interactive one gave an abbreviated object name).
51
64
52
- - "git send-email" did not unquote encoded words that appear on the
65
+ * "git send-email" did not unquote encoded words that appear on the
53
66
header correctly, and lost "_" from strings.
54
67
55
- - When the user gives an argument that can be taken as both a
56
- revision name and a pathname without disambiguating with "--", we
57
- used to give a help message "Use '--' to separate". The message
58
- has been clarified to show where that '--' goes on the command
59
- line .
68
+ * "git stash apply/pop" did not trigger "rerere" upon conflicts
69
+ unlike other mergy operations.
70
+
71
+ * "git submodule <cmd> path" did not error out when the path to the
72
+ submodule was misspelt .
60
73
61
- - "gitweb" when used with PATH_INFO failed to notice directories with
74
+ * "git submodule update -f" did not update paths in the working tree
75
+ that has local changes.
76
+ (merge 01d4721 sz/submodule-force-update later to maint).
77
+
78
+ * "gitweb" when used with PATH_INFO failed to notice directories with
62
79
SP (and other characters that need URL-style quoting) in them.
80
+
81
+ * Fallback 'getpass' implementation made unportable use of stdio API.
82
+
83
+ * A utility shell function test_seq has been added as a replacement
84
+ for the 'seq' utility found on some platforms.
0 commit comments