Sync with 1.7.11-rc0

By Junio C Hamano (2) and others
* master:
  Git 1.7.11-rc0
  Git 1.7.10.3
  fmt-merge-message: add empty line between tag and signature verification
  config doc: remove confusion about relative GIT_DIR from FILES section

Documentation/RelNotes/1.7.10.3.txt

@@ -6,10 +6,6 @@ Fixes since v1.7.10.2
 
  * The message file for German translation has been updated a bit.
 
- * "git status --porcelain" ignored "--branch" option by mistake.  The
-   output for "git status --branch -z" was also incorrect and did not
-   terminate the record for the current branch name with NUL as asked.
-
  * Running "git checkout" on an unborn branch used to corrupt HEAD.
 
  * When checking out another commit from an already detached state, we
@@ -21,6 +17,9 @@ Fixes since v1.7.10.2
    "checkout" phase; when run without any "--quiet" option, it should
    give progress to the lengthy operation.
 
+ * The directory path used in "git diff --no-index", when it recurses
+   down, was broken with a recent update after v1.7.10.1 release.
+
  * "log -z --pretty=tformat:..." did not terminate each record with
    NUL.  The fix is not entirely correct when the output also asks for
    --patch and/or --stat, though.
@@ -29,4 +28,16 @@ Fixes since v1.7.10.2
    broken and gave undue precedence to configured log.date, causing
    "git stash list" to show "stash@{time stamp string}".
 
+ * "git status --porcelain" ignored "--branch" option by mistake.  The
+   output for "git status --branch -z" was also incorrect and did not
+   terminate the record for the current branch name with NUL as asked.
+
+ * When a submodule repository uses alternate object store mechanism,
+   some commands that were started from the superproject did not
+   notice it and failed with "No such object" errors.  The subcommands
+   of "git submodule" command that recursed into the submodule in a
+   separate process were OK; only the ones that cheated and peeked
+   directly into the submodule's repository from the primary process
+   were affected.
+
 Also contains minor fixes and documentation updates.

Documentation/RelNotes/1.7.11.txt

@@ -79,6 +79,9 @@ Foreign Interface
 
 Performance and Internal Implementation (please report possible regressions)
 
+ * Bash completion script (in contrib/) have been cleaned up to make
+   future work on it simpler.
+
  * An experimental "version 4" format of the index file has been
    introduced to reduce on-disk footprint and I/O overhead.
 
@@ -120,23 +123,26 @@ Unless otherwise noted, all the fixes since v1.7.10 in the maintenance
 releases are contained in this release (see release notes to them for
 details).
 
- * When a submodule repository uses alternate object store mechanism,
+ * The progress indicator for a large "git checkout" was sent to
-   some commands that were started from the superproject did not
+   stderr even if it is not a terminal.
-   notice it and failed with "No such object" errors.  The subcommands
+   (merge e9fc64c ap/checkout-no-progress-for-non-tty later to maint).
-   of "git submodule" command that properly recursed into the
-   submodule in a separate process were OK; only the ones that cheated
-   and peeked directly into the submodule's repository from the
-   primary process were affected.
-   (merge 5e73633 hv/submodule-alt-odb later to maint).
 
- * The directory path used in "git diff --no-index", when it recurses
+ * A name taken from mailmap was copied into an internal buffer
-   down, was broken with a recent update after v1.7.10.1 release.
+   incorrectly and could overun the buffer if it is too long.
-   (merge 176a335 bp/diff-no-index-strbuf-fix later to maint).
+   (merge c9b4e9e jk/format-person-part-buffer-limit later to maint).
 
- * "git status --porcelain" ignored "--branch" option by mistake.  The
+ * A malformed commit object that has a header line chomped in the
-   output for "git status --branch -z" was also incorrect and did not
+   middle could kill git with a NULL pointer dereference.
-   terminate the record for the current branch name with NUL as asked.
+   (merge a9c7a8a jk/pretty-commit-header-incomplete-line later to maint).
-   (merge d4a6bf1 jk/maint-status-porcelain-z-b later to maint).
+
+ * An author/committer name that is a single character was mishandled
+   as an invalid name by mistake.
+   (merge d9955fd jk/ident-split-fix later to maint).
+
+ * "git grep -e '$pattern'", unlike the case where the patterns are
+   read from a file, did not treat individual lines in the given
+   pattern argument as separate regular expressions as it should.
+   (merge ec83061 rs/maint-grep-F later to maint).
 
  * "git diff --stat" used to fully count a binary file with modified
    execution bits whose contents is unmodified, which was not quite

Documentation/git-config.txt

@@ -198,9 +198,7 @@ If not set explicitly with '--file', there are three files where
 'git config' will search for configuration options:
 
 $GIT_DIR/config::
-	Repository specific configuration file. (The filename is
+	Repository specific configuration file.
-	of course relative to the repository root, not the working
-	directory.)
 
 ~/.gitconfig::
 	User-specific configuration file. Also called "global"

GIT-VERSION-GEN

@@ -1,7 +1,7 @@
 #!/bin/sh
 
 GVF=GIT-VERSION-FILE
-DEF_VER=v1.7.10.GIT
+DEF_VER=v1.7.11-rc0
 
 LF='
 '

builtin/fmt-merge-msg.c

@@ -462,7 +462,10 @@ static void fmt_tag_signature(struct strbuf *tagbuf,
 		strbuf_add(tagbuf, tag_body, buf + len - tag_body);
 	}
 	strbuf_complete_line(tagbuf);
-	strbuf_add_lines(tagbuf, "# ", sig->buf, sig->len);
+	if (sig->len) {
+		strbuf_addch(tagbuf, '\n');
+		strbuf_add_lines(tagbuf, "# ", sig->buf, sig->len);
+	}
 }
 
 static void fmt_merge_msg_sigs(struct strbuf *out)
@@ -627,8 +630,7 @@ int fmt_merge_msg(struct strbuf *in, struct strbuf *out,
 		rev.ignore_merges = 1;
 		rev.limited = 1;
 
-		if (suffixcmp(out->buf, "\n"))
+		strbuf_complete_line(out);
-			strbuf_addch(out, '\n');
 
 		for (i = 0; i < origins.nr; i++)
 			shortlog(origins.items[i].string,