In his latest "What's cooking in git.git" message (Nov 2011, msysgit#2; Sun, 6)
J.C.Hamano announced the availability of the documentation
branches (html, doc) in dedicated repositories at github.com.
Change .gitmodules to point 'doc/git/html' to the new repository.

Signed-off-by: Stefan Naewe <[email protected]>
Signed-off-by: Pat Thoyts <[email protected]>

We avoid setting `encoding' and `termencoding', and opt to only setting
`fileencodings' because it's understood that vim sets the former two to
the user's system's "ANSI" and "OEM" code pages (setting either both if
they differ or only `encoding' if they're the same) [1].  The setting of
"OEM" is particularly significant as it is required for native characters
to be correctly displayed on the user's console [2].

We set "utf-8" as the only value of `fileencodings', there by making it
the default encoding used to save new commit messages to COMMIT_EDITMSG.
This is consistent with what git expects in an environment where the
user has not altered the value of `i18n.commitencoding'.

When loading an existing message, for operations such as 'commit --amend'
and 'rebase -i', our setting for `fileencodings' means that vim will first
attempt to load it as UTF-8 and then fallback to the encoding set by
`encoding' if unable to do so.  This encoding is expected to be "ANSI",
as mentioned above, and this side-effect is of benefit if the user does
have an altered value of `i18n.commitencoding'.  (Here, we're assuming
users not set this, if at all, to anything other than that of "ANSI"
because that is thought to be the most natural for their location.)

Users SHOULD unset `i18n.commitencoding' after this is released.  However,
failing to do so may not show any visible regressions.  (See tests for
case msysgit#2 at [3].)

This patch follows the discussion in the [4] link.

[1] http://groups.google.com/group/msysgit/msg/be6218f1cbf5e747
[2] http://groups.google.com/group/vim_dev/msg/3dd77f3014a60e43
[3] http://groups.google.com/group/msysgit/msg/508cca8b57f00546
[4] http://groups.google.com/group/msysgit/msg/55c582fb1b279d08

"git add -u" updates the index with the updated contents from the working
tree by internally running "diff-files" to grab the set of paths that are
different from the index. Then it updates the index entries for the paths
that are modified in the working tree, and deletes the index entries for
the paths that are deleted in the working tree.

It ignored the output from the diff-files that indicated that a path is
unmerged.  For these paths, it instead relied on the fact that an unmerged
path is followed by the result of comparison between stage msysgit#2 (ours) and
the working tree, and used that to update or delete such a path when it is
used to record the resolution of a conflict.

As the result, when a path did not have stage msysgit#2 (e.g. "we deleted while
the other side added"), these unmerged stages were left behind, instead of
recording what the user resolved in the working tree.

Since we recently fixed "diff-files" to indicate if the corresponding path
exists on the working tree for an unmerged path, we do not have to rely on
the comparison with stage msysgit#2 anymore. We can instead tell the diff-files
not to compare with higher stages, and use the unmerged output to update
the index to reflect the state of the working tree.

The changes to the test vector in t2200 illustrates the nature of the bug
and the fix.  The test expected stage msysgit#1 and msysgit#3 entries be left behind,
but it was codifying the buggy behaviour.

Signed-off-by: Junio C Hamano <[email protected]>

The previous logic in show_config was to print the delimiter when the
value was set, but Boolean variables have an implicit value "true" when
they appear with no value in the config file. As a result, we got:

git_Config        --get-regexp '.*\.Boolean'	msysgit#1. Ok: example.boolean
git_Config --bool --get-regexp '.*\.Boolean'	msysgit#2. NO: example.booleantrue

Fix this by defering the display of the separator until after the value
to display has been computed.

Reported-by: Brian Foster <[email protected]>
Signed-off-by: Matthieu Moy <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>

While identifying the commit merged to our history as "parent msysgit#2" is
technically correct, we will never say "parent msysgit#1" (as that is the tip of
our history before the merge is made), and we rarely would say "parent msysgit#3"
(which would mean the merge is an octopus), especially when responding to
a request to pull a signed tag.

Treat the most common case to merge a single commit specially, and just
say "merged tag '<tagname>'" instead.

Signed-off-by: Junio C Hamano <[email protected]>

Even though we show a separate *UNMERGED* entry in the patch and
diffstat output (or in the --raw format, for that matter) in
addition to and separately from the diff against the specified stage
(defaulting to msysgit#2) for unmerged paths, they should not be counted in
the total number of files affected---that would lead to counting the
same path twice.

The separation done by the previous step makes this fix simple and
straightforward.  Among the filepairs in diff_queue, paths that
weren't modified, and the extra "unmerged" entries do not count as
total number of files.

Signed-off-by: Junio C Hamano <[email protected]>

entry_count is used in update_one() for two purposes:

1. to skip through the number of processed entries in in-memory index
2. to record the number of entries this cache-tree covers on disk

Unfortunately when CE_REMOVE is present these numbers are not the same
because CE_REMOVE entries are automatically removed before writing to
disk but entry_count is not adjusted and still counts CE_REMOVE
entries.

Separate the two use cases into two different variables. msysgit#1 is taken
care by the new field count in struct cache_tree_sub and entry_count
is prepared for msysgit#2.

Signed-off-by: Nguyễn Thái Ngọc Duy <[email protected]>
Signed-off-by: Junio C Hamano <[email protected]>

Closing (not redirecting to /dev/null) the standard error stream is
not a very smart thing to do.  Later open may return file
descriptor msysgit#2 for unrelated purpose, and error reporting code may
write into them.

* tr/perl-keep-stderr-open:
  t9700: do not close STDERR
  perl: redirect stderr to /dev/null instead of closing