提交 04f42624 编写于 作者: J Junio C Hamano

Sync with maint

Signed-off-by: NJunio C Hamano <gitster@pobox.com>
...@@ -4,59 +4,81 @@ Git v1.7.11.6 Release Notes ...@@ -4,59 +4,81 @@ Git v1.7.11.6 Release Notes
Fixes since v1.7.11.5 Fixes since v1.7.11.5
--------------------- ---------------------
This consists primarily of documentation updates and low-impact code * "ciabot" script (in contrib/) has been updated with extensive
clarification and bugfixes.
- "ciabot" script (in contrib/) has been updated with extensive
documentation. documentation.
- The "--rebase" option to "git pull" can be abbreviated to "-r", * "git foo" errored out with "Not a directory" when the user had a
but we didn't document it. non-directory on $PATH, and worse yet it masked an alias "foo" from
running.
- It was generally understood that "--long-option"s to many of our * When the user exports a non-default IFS without HT, scripts that
subcommands can be abbreviated to the unique prefix, but it was not rely on being able to parse "ls-files -s | while read a b c..."
easy to find it described for new readers of the documentation set. started to fail. Protect them from such a misconfiguration.
- The "--topo-order", "--date-order" (and the lack of either means * When the user gives an argument that can be taken as both a
the default order) options to "rev-list" and "log" family of revision name and a pathname without disambiguating with "--", we
commands were poorly described in the documentation. used to give a help message "Use '--' to separate". The message
has been clarified to show where that '--' goes on the command
line.
* Documentation for the configuration file format had a confusing
example.
- Older parts of the documentation described as if having a regular * Older parts of the documentation described as if having a regular
file in .git/refs/ hierarchy were the only way to have branches and file in .git/refs/ hierarchy were the only way to have branches and
tags, which is not true for quite some time. tags, which is not true for quite some time.
- A utility shell function test_seq has been added as a replacement * It was generally understood that "--long-option"s to many of our
for the 'seq' utility found on some platforms. subcommands can be abbreviated to the unique prefix, but it was not
easy to find it described for new readers of the documentation set.
- Fallback 'getpass' implementation made unportable use of stdio API. * The "--topo-order", "--date-order" (and the lack of either means
the default order) options to "rev-list" and "log" family of
commands were poorly described in the documentation.
- "git commit --amend" let the user edit the log message and then * "git commit --amend" let the user edit the log message and then
died when the human-readable committer name was given died when the human-readable committer name was given
insufficiently by getpwent(3). insufficiently by getpwent(3).
- The reflog entries left by "git rebase" and "git rebase -i" were * The exit status code from "git config" was way overspecified while
inconsistent (the interactive one gave an abbreviated object name). being incorrect. The implementation has been updated to give the
documented status for a case that was documented, and introduce a
new code for "all other errors".
- When the user exports a non-default IFS without HT, scripts that * The output from "git diff -B" for a file that ends with an
rely on being able to parse "ls-files -s | while read a b c..." incomplete line did not put "\ No newline..." on a line of its own.
started to fail. Protect them from such a misconfiguration.
* "git diff" had a confusion between taking data from a path in the
working tree and taking data from an object that happens to have
name 0{40} recorded in a tree.
- When "git push" triggered the automatic gc on the receiving end, a * The "--rebase" option to "git pull" can be abbreviated to "-r",
but we didn't document it.
* When "git push" triggered the automatic gc on the receiving end, a
message from "git prune" that said it was removing cruft leaked to message from "git prune" that said it was removing cruft leaked to
the standard output, breaking the communication protocol. the standard output, breaking the communication protocol.
- "git diff" had a confusion between taking data from a path in the * The reflog entries left by "git rebase" and "git rebase -i" were
working tree and taking data from an object that happens to have inconsistent (the interactive one gave an abbreviated object name).
name 0{40} recorded in a tree.
- "git send-email" did not unquote encoded words that appear on the * "git send-email" did not unquote encoded words that appear on the
header correctly, and lost "_" from strings. header correctly, and lost "_" from strings.
- When the user gives an argument that can be taken as both a * "git stash apply/pop" did not trigger "rerere" upon conflicts
revision name and a pathname without disambiguating with "--", we unlike other mergy operations.
used to give a help message "Use '--' to separate". The message
has been clarified to show where that '--' goes on the command * "git submodule <cmd> path" did not error out when the path to the
line. submodule was misspelt.
- "gitweb" when used with PATH_INFO failed to notice directories with * "git submodule update -f" did not update paths in the working tree
that has local changes.
(merge 01d4721 sz/submodule-force-update later to maint).
* "gitweb" when used with PATH_INFO failed to notice directories with
SP (and other characters that need URL-style quoting) in them. SP (and other characters that need URL-style quoting) in them.
* Fallback 'getpass' implementation made unportable use of stdio API.
* A utility shell function test_seq has been added as a replacement
for the 'seq' utility found on some platforms.
...@@ -25,12 +25,25 @@ Fixes since v1.7.12 ...@@ -25,12 +25,25 @@ Fixes since v1.7.12
* A utility shell function test_seq has been added as a replacement * A utility shell function test_seq has been added as a replacement
for the 'seq' utility found on some platforms. for the 'seq' utility found on some platforms.
* Compatibility wrapper to learn the maximum number of file
descriptors we can open around sysconf(_SC_OPEN_MAX) and
getrlimit(RLIMIT_NO_FILE) has been introduced for portability.
* We used curl_easy_strerror() without checking version of cURL,
breaking the build for versions before curl 7.12.0.
* Code to work around MacOS X UTF-8 gotcha has been cleaned up.
* Fallback 'getpass' implementation made unportable use of stdio API. * Fallback 'getpass' implementation made unportable use of stdio API.
* "git commit --amend" let the user edit the log message and then * "git commit --amend" let the user edit the log message and then
died when the human-readable committer name was given died when the human-readable committer name was given
insufficiently by getpwent(3). insufficiently by getpwent(3).
* Some capabilities were asked by fetch-pack even when upload-pack
did not advertise that they are available. fetch-pack has been
fixed not to do so.
* "git prune" without "-v" used to warn about leftover temporary * "git prune" without "-v" used to warn about leftover temporary
files (which is an indication of an earlier aborted operation). files (which is an indication of an earlier aborted operation).
......
...@@ -48,9 +48,10 @@ Documentation for older releases are available here: ...@@ -48,9 +48,10 @@ Documentation for older releases are available here:
* release notes for * release notes for
link:RelNotes/1.7.12.txt[1.7.12]. link:RelNotes/1.7.12.txt[1.7.12].
* link:v1.7.11.5/git.html[documentation for release 1.7.11.5] * link:v1.7.11.6/git.html[documentation for release 1.7.11.6]
* release notes for * release notes for
link:RelNotes/1.7.11.6.txt[1.7.11.6],
link:RelNotes/1.7.11.5.txt[1.7.11.5], link:RelNotes/1.7.11.5.txt[1.7.11.5],
link:RelNotes/1.7.11.4.txt[1.7.11.4], link:RelNotes/1.7.11.4.txt[1.7.11.4],
link:RelNotes/1.7.11.3.txt[1.7.11.3], link:RelNotes/1.7.11.3.txt[1.7.11.3],
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册