Merge branch 'as/check-ignore'

Enhance "check-ignore" (1.8.2 update) to work more like "check-attr"
over bidi-pipes.

* as/check-ignore:
  t0008: use named pipe (FIFO) to test check-ignore streaming
  Documentation: add caveats about I/O buffering for check-{attr,ignore}
  check-ignore: allow incremental streaming of queries via --stdin
  check-ignore: move setup into cmd_check_ignore()
  check-ignore: add -n / --non-matching option
  t0008: remove duplicated test fixture data

5 files changed, 188 insertions, 96 deletions

diff --git a/Documentation/git-check-attr.txt b/Documentation/git-check-attr.txt
index 5abdbaa51..a7be80d48 100644
--- a/Documentation/git-check-attr.txt
+++ b/Documentation/git-check-attr.txt
@@ -56,6 +56,11 @@ being queried and <info> can be either:
 'set';;		when the attribute is defined as true.
 <value>;;	when a value has been assigned to the attribute.
 
+Buffering happens as documented under the `GIT_FLUSH` option in
+linkgit:git[1].  The caller is responsible for avoiding deadlocks
+caused by overfilling an input buffer or reading from an empty output
+buffer.
+
 EXAMPLES
 --------
 
diff --git a/Documentation/git-check-ignore.txt b/Documentation/git-check-ignore.txt
index 854e4d0c4..8e1f7ab7e 100644
--- a/Documentation/git-check-ignore.txt
+++ b/Documentation/git-check-ignore.txt
@@ -39,6 +39,12 @@ OPTIONS
 	below).  If `--stdin` is also given, input paths are separated
 	with a NUL character instead of a linefeed character.
 
+-n, --non-matching::
+	Show given paths which don't match any pattern.	 This only
+	makes sense when `--verbose` is enabled, otherwise it would
+	not be possible to distinguish between paths which match a
+	pattern and those which don't.
+
 OUTPUT
 ------
 
@@ -65,6 +71,20 @@ are also used instead of colons and hard tabs:
 
 <source> <NULL> <linenum> <NULL> <pattern> <NULL> <pathname> <NULL>
 
+If `-n` or `--non-matching` are specified, non-matching pathnames will
+also be output, in which case all fields in each output record except
+for <pathname> will be empty.  This can be useful when running
+non-interactively, so that files can be incrementally streamed to
+STDIN of a long-running check-ignore process, and for each of these
+files, STDOUT will indicate whether that file matched a pattern or
+not.  (Without this option, it would be impossible to tell whether the
+absence of output for a given file meant that it didn't match any
+pattern, or that the output hadn't been generated yet.)
+
+Buffering happens as documented under the `GIT_FLUSH` option in
+linkgit:git[1].  The caller is responsible for avoiding deadlocks
+caused by overfilling an input buffer or reading from an empty output
+buffer.
 
 EXIT STATUS
 -----------
diff --git a/Documentation/git.txt b/Documentation/git.txt
index 9e302b0a6..65de534e4 100644
--- a/Documentation/git.txt
+++ b/Documentation/git.txt
@@ -811,8 +811,9 @@ for further details.
 'GIT_FLUSH'::
 	If this environment variable is set to "1", then commands such
 	as 'git blame' (in incremental mode), 'git rev-list', 'git log',
-	and 'git whatchanged' will force a flush of the output stream
-	after each commit-oriented record have been flushed.   If this
+	'git check-attr', 'git check-ignore', and 'git whatchanged' will
+	force a flush of the output stream after each record have been
+	flushed. If this
 	variable is set to "0", the output of these commands will be done
 	using completely buffered I/O.   If this environment variable is
 	not set, Git will choose buffered or record-oriented flushing
diff --git a/builtin/check-ignore.c b/builtin/check-ignore.c
index 854a88a05..4a8fc707c 100644
--- a/builtin/check-ignore.c
+++ b/builtin/check-ignore.c
@@ -5,7 +5,7 @@
 #include "pathspec.h"
 #include "parse-options.h"
 
-static int quiet, verbose, stdin_paths;
+static int quiet, verbose, stdin_paths, show_non_matching;
 static const char * const check_ignore_usage[] = {
 "git check-ignore [options] pathname...",
 "git check-ignore [options] --stdin < <list-of-paths>",
@@ -22,21 +22,28 @@ static const struct option check_ignore_options[] = {
 		    N_("read file names from stdin")),
 	OPT_BOOLEAN('z', NULL, &null_term_line,
 		    N_("input paths are terminated by a null character")),
+	OPT_BOOLEAN('n', "non-matching", &show_non_matching,
+		    N_("show non-matching input paths")),
 	OPT_END()
 };
 
 static void output_exclude(const char *path, struct exclude *exclude)
 {
-	char *bang  = exclude->flags & EXC_FLAG_NEGATIVE  ? "!" : "";
-	char *slash = exclude->flags & EXC_FLAG_MUSTBEDIR ? "/" : "";
+	char *bang  = (exclude && exclude->flags & EXC_FLAG_NEGATIVE)  ? "!" : "";
+	char *slash = (exclude && exclude->flags & EXC_FLAG_MUSTBEDIR) ? "/" : "";
 	if (!null_term_line) {
 		if (!verbose) {
 			write_name_quoted(path, stdout, '\n');
 		} else {
-			quote_c_style(exclude->el->src, NULL, stdout, 0);
-			printf(":%d:%s%s%s\t",
-			       exclude->srcpos,
-			       bang, exclude->pattern, slash);
+			if (exclude) {
+				quote_c_style(exclude->el->src, NULL, stdout, 0);
+				printf(":%d:%s%s%s\t",
+				       exclude->srcpos,
+				       bang, exclude->pattern, slash);
+			}
+			else {
+				printf("::\t");
+			}
 			quote_c_style(path, NULL, stdout, 0);
 			fputc('\n', stdout);
 		}
@@ -44,30 +51,26 @@ static void output_exclude(const char *path, struct exclude *exclude)
 		if (!verbose) {
 			printf("%s%c", path, '\0');
 		} else {
-			printf("%s%c%d%c%s%s%s%c%s%c",
-			       exclude->el->src, '\0',
-			       exclude->srcpos, '\0',
-			       bang, exclude->pattern, slash, '\0',
-			       path, '\0');
+			if (exclude)
+				printf("%s%c%d%c%s%s%s%c%s%c",
+				       exclude->el->src, '\0',
+				       exclude->srcpos, '\0',
+				       bang, exclude->pattern, slash, '\0',
+				       path, '\0');
+			else
+				printf("%c%c%c%s%c", '\0', '\0', '\0', path, '\0');
 		}
 	}
 }
 
-static int check_ignore(const char *prefix, const char **pathspec)
+static int check_ignore(struct dir_struct *dir,
+			const char *prefix, const char **pathspec)
 {
-	struct dir_struct dir;
 	const char *path, *full_path;
 	char *seen;
 	int num_ignored = 0, dtype = DT_UNKNOWN, i;
 	struct exclude *exclude;
 
-	/* read_cache() is only necessary so we can watch out for submodules. */
-	if (read_cache() < 0)
-		die(_("index file corrupt"));
-
-	memset(&dir, 0, sizeof(dir));
-	setup_standard_excludes(&dir);
-
 	if (!pathspec || !*pathspec) {
 		if (!quiet)
 			fprintf(stderr, "no pathspec given.\n");
@@ -86,28 +89,26 @@ static int check_ignore(const char *prefix, const char **pathspec)
 					? strlen(prefix) : 0, path);
 		full_path = check_path_for_gitlink(full_path);
 		die_if_path_beyond_symlink(full_path, prefix);
+		exclude = NULL;
 		if (!seen[i]) {
-			exclude = last_exclude_matching(&dir, full_path, &dtype);
-			if (exclude) {
-				if (!quiet)
-					output_exclude(path, exclude);
-				num_ignored++;
-			}
+			exclude = last_exclude_matching(dir, full_path, &dtype);
 		}
+		if (!quiet && (exclude || show_non_matching))
+			output_exclude(path, exclude);
+		if (exclude)
+			num_ignored++;
 	}
 	free(seen);
-	clear_directory(&dir);
 
 	return num_ignored;
 }
 
-static int check_ignore_stdin_paths(const char *prefix)
+static int check_ignore_stdin_paths(struct dir_struct *dir, const char *prefix)
 {
 	struct strbuf buf, nbuf;
-	char **pathspec = NULL;
-	size_t nr = 0, alloc = 0;
+	char *pathspec[2] = { NULL, NULL };
 	int line_termination = null_term_line ? 0 : '\n';
-	int num_ignored;
+	int num_ignored = 0;
 
 	strbuf_init(&buf, 0);
 	strbuf_init(&nbuf, 0);
@@ -118,23 +119,19 @@ static int check_ignore_stdin_paths(const char *prefix)
 				die("line is badly quoted");
 			strbuf_swap(&buf, &nbuf);
 		}
-		ALLOC_GROW(pathspec, nr + 1, alloc);
-		pathspec[nr] = xcalloc(strlen(buf.buf) + 1, sizeof(*buf.buf));
-		strcpy(pathspec[nr++], buf.buf);
+		pathspec[0] = buf.buf;
+		num_ignored += check_ignore(dir, prefix, (const char **)pathspec);
+		maybe_flush_or_die(stdout, "check-ignore to stdout");
 	}
-	ALLOC_GROW(pathspec, nr + 1, alloc);
-	pathspec[nr] = NULL;
-	num_ignored = check_ignore(prefix, (const char **)pathspec);
-	maybe_flush_or_die(stdout, "attribute to stdout");
 	strbuf_release(&buf);
 	strbuf_release(&nbuf);
-	free(pathspec);
 	return num_ignored;
 }
 
 int cmd_check_ignore(int argc, const char **argv, const char *prefix)
 {
 	int num_ignored;
+	struct dir_struct dir;
 
 	git_config(git_default_config, NULL);
 
@@ -156,13 +153,24 @@ int cmd_check_ignore(int argc, const char **argv, const char *prefix)
 		if (verbose)
 			die(_("cannot have both --quiet and --verbose"));
 	}
+	if (show_non_matching && !verbose)
+		die(_("--non-matching is only valid with --verbose"));
+
+	/* read_cache() is only necessary so we can watch out for submodules. */
+	if (read_cache() < 0)
+		die(_("index file corrupt"));
+
+	memset(&dir, 0, sizeof(dir));
+	setup_standard_excludes(&dir);
 
 	if (stdin_paths) {
-		num_ignored = check_ignore_stdin_paths(prefix);
+		num_ignored = check_ignore_stdin_paths(&dir, prefix);
 	} else {
-		num_ignored = check_ignore(prefix, argv);
+		num_ignored = check_ignore(&dir, prefix, argv);
 		maybe_flush_or_die(stdout, "ignore to stdout");
 	}
 
+	clear_directory(&dir);
+
 	return !num_ignored;
 }
diff --git a/t/t0008-ignores.sh b/t/t0008-ignores.sh
index 9c1bde1fd..a56db804c 100755
--- a/t/t0008-ignores.sh
+++ b/t/t0008-ignores.sh
@@ -66,16 +66,23 @@ test_check_ignore () {
 
 	init_vars &&
 	rm -f "$HOME/stdout" "$HOME/stderr" "$HOME/cmd" &&
-	echo git $global_args check-ignore $quiet_opt $verbose_opt $args \
+	echo git $global_args check-ignore $quiet_opt $verbose_opt $non_matching_opt $args \
 		>"$HOME/cmd" &&
+	echo "$expect_code" >"$HOME/expected-exit-code" &&
 	test_expect_code "$expect_code" \
-		git $global_args check-ignore $quiet_opt $verbose_opt $args \
+		git $global_args check-ignore $quiet_opt $verbose_opt $non_matching_opt $args \
 		>"$HOME/stdout" 2>"$HOME/stderr" &&
 	test_cmp "$HOME/expected-stdout" "$HOME/stdout" &&
 	stderr_empty_on_success "$expect_code"
 }
 
-# Runs the same code with 3 different levels of output verbosity,
+# Runs the same code with 4 different levels of output verbosity:
+#
+#   1. with -q / --quiet
+#   2. with default verbosity
+#   3. with -v / --verbose
+#   4. with -v / --verbose, *and* -n / --non-matching
+#
 # expecting success each time.  Takes advantage of the fact that
 # check-ignore --verbose output is the same as normal output except
 # for the extra first column.
@@ -83,7 +90,9 @@ test_check_ignore () {
 # Arguments:
 #   - (optional) prereqs for this test, e.g. 'SYMLINKS'
 #   - test name
-#   - output to expect from -v / --verbose mode
+#   - output to expect from the fourth verbosity mode (the output
+#     from the other verbosity modes is automatically inferred
+#     from this value)
 #   - code to run (should invoke test_check_ignore)
 test_expect_success_multi () {
 	prereq=
@@ -92,8 +101,9 @@ test_expect_success_multi () {
 		prereq=$1
 		shift
 	fi
-	testname="$1" expect_verbose="$2" code="$3"
+	testname="$1" expect_all="$2" code="$3"
 
+	expect_verbose=$( echo "$expect_all" | grep -v '^::	' )
 	expect=$( echo "$expect_verbose" | sed -e 's/.*	//' )
 
 	test_expect_success $prereq "$testname" '
@@ -101,23 +111,40 @@ test_expect_success_multi () {
 		eval "$code"
 	'
 
-	for quiet_opt in '-q' '--quiet'
-	do
-		test_expect_success $prereq "$testname${quiet_opt:+ with $quiet_opt}" "
+	# --quiet is only valid when a single pattern is passed
+	if test $( echo "$expect_all" | wc -l ) = 1
+	then
+		for quiet_opt in '-q' '--quiet'
+		do
+			test_expect_success $prereq "$testname${quiet_opt:+ with $quiet_opt}" "
 			expect '' &&
 			$code
 		"
-	done
-	quiet_opt=
+		done
+		quiet_opt=
+	fi
 
 	for verbose_opt in '-v' '--verbose'
 	do
-		test_expect_success $prereq "$testname${verbose_opt:+ with $verbose_opt}" "
-			expect '$expect_verbose' &&
-			$code
-		"
+		for non_matching_opt in '' ' -n' ' --non-matching'
+		do
+			if test -n "$non_matching_opt"
+			then
+				my_expect="$expect_all"
+			else
+				my_expect="$expect_verbose"
+			fi
+
+			test_code="
+				expect '$my_expect' &&
+				$code
+			"
+			opts="$verbose_opt$non_matching_opt"
+			test_expect_success $prereq "$testname${opts:+ with $opts}" "$test_code"
+		done
 	done
 	verbose_opt=
+	non_matching_opt=
 }
 
 test_expect_success 'setup' '
@@ -178,7 +205,7 @@ test_expect_success 'setup' '
 #
 # test invalid inputs
 
-test_expect_success_multi '. corner-case' '' '
+test_expect_success_multi '. corner-case' '::	.' '
 	test_check_ignore . 1
 '
 
@@ -189,11 +216,7 @@ test_expect_success_multi 'empty command line' '' '
 
 test_expect_success_multi '--stdin with empty STDIN' '' '
 	test_check_ignore "--stdin" 1 </dev/null &&
-	if test -n "$quiet_opt"; then
-		test_stderr ""
-	else
-		test_stderr "no pathspec given."
-	fi
+	test_stderr ""
 '
 
 test_expect_success '-q with multiple args' '
@@ -276,27 +299,39 @@ do
 		where="in subdir $subdir"
 	fi
 
-	test_expect_success_multi "non-existent file $where not ignored" '' "
-		test_check_ignore '${subdir}non-existent' 1
-	"
+	test_expect_success_multi "non-existent file $where not ignored" \
+		"::	${subdir}non-existent" \
+		"test_check_ignore '${subdir}non-existent' 1"
 
 	test_expect_success_multi "non-existent file $where ignored" \
-		".gitignore:1:one	${subdir}one" "
-		test_check_ignore '${subdir}one'
-	"
+		".gitignore:1:one	${subdir}one" \
+		"test_check_ignore '${subdir}one'"
 
-	test_expect_success_multi "existing untracked file $where not ignored" '' "
-		test_check_ignore '${subdir}not-ignored' 1
-	"
+	test_expect_success_multi "existing untracked file $where not ignored" \
+		"::	${subdir}not-ignored" \
+		"test_check_ignore '${subdir}not-ignored' 1"
 
-	test_expect_success_multi "existing tracked file $where not ignored" '' "
-		test_check_ignore '${subdir}ignored-but-in-index' 1
-	"
+	test_expect_success_multi "existing tracked file $where not ignored" \
+		"::	${subdir}ignored-but-in-index" \
+		"test_check_ignore '${subdir}ignored-but-in-index' 1"
 
 	test_expect_success_multi "existing untracked file $where ignored" \
-		".gitignore:2:ignored-*	${subdir}ignored-and-untracked" "
-		test_check_ignore '${subdir}ignored-and-untracked'
-	"
+		".gitignore:2:ignored-*	${subdir}ignored-and-untracked" \
+		"test_check_ignore '${subdir}ignored-and-untracked'"
+
+	test_expect_success_multi "mix of file types $where" \
+"::	${subdir}non-existent
+.gitignore:1:one	${subdir}one
+::	${subdir}not-ignored
+::	${subdir}ignored-but-in-index
+.gitignore:2:ignored-*	${subdir}ignored-and-untracked" \
+		"test_check_ignore '
+			${subdir}non-existent
+			${subdir}one
+			${subdir}not-ignored
+			${subdir}ignored-but-in-index
+			${subdir}ignored-and-untracked'
+		"
 done
 
 # Having established the above, from now on we mostly test against
@@ -391,7 +426,7 @@ test_expect_success 'cd to ignored sub-directory with -v' '
 #
 # test handling of symlinks
 
-test_expect_success_multi SYMLINKS 'symlink' '' '
+test_expect_success_multi SYMLINKS 'symlink' '::	a/symlink' '
 	test_check_ignore "a/symlink" 1
 '
 
@@ -574,37 +609,34 @@ cat <<-\EOF >stdin
 	globaltwo
 	b/globaltwo
 	../b/globaltwo
+	c/not-ignored
 EOF
-cat <<-\EOF >expected-default
-	../one
-	one
-	b/on
-	b/one
-	b/one one
-	b/one two
-	"b/one\"three"
-	b/two
-	b/twooo
-	../globaltwo
-	globaltwo
-	b/globaltwo
-	../b/globaltwo
-EOF
-cat <<-EOF >expected-verbose
+# N.B. we deliberately end STDIN with a non-matching pattern in order
+# to test that the exit code indicates that one or more of the
+# provided paths is ignored - in other words, that it represents an
+# aggregation of all the results, not just the final result.
+
+cat <<-EOF >expected-all
 	.gitignore:1:one	../one
+	::	../not-ignored
 	.gitignore:1:one	one
+	::	not-ignored
 	a/b/.gitignore:8:!on*	b/on
 	a/b/.gitignore:8:!on*	b/one
 	a/b/.gitignore:8:!on*	b/one one
 	a/b/.gitignore:8:!on*	b/one two
 	a/b/.gitignore:8:!on*	"b/one\"three"
 	a/b/.gitignore:9:!two	b/two
+	::	b/not-ignored
 	a/.gitignore:1:two*	b/twooo
 	$global_excludes:2:!globaltwo	../globaltwo
 	$global_excludes:2:!globaltwo	globaltwo
 	$global_excludes:2:!globaltwo	b/globaltwo
 	$global_excludes:2:!globaltwo	../b/globaltwo
+	::	c/not-ignored
 EOF
+grep -v '^::	' expected-all >expected-verbose
+sed -e 's/.*	//' expected-verbose >expected-default
 
 sed -e 's/^"//' -e 's/\\//' -e 's/"$//' stdin | \
 	tr "\n" "\0" >stdin0
@@ -629,6 +661,14 @@ test_expect_success '--stdin from subdirectory with -v' '
 	)
 '
 
+test_expect_success '--stdin from subdirectory with -v -n' '
+	expect_from_stdin <expected-all &&
+	(
+		cd a &&
+		test_check_ignore "--stdin -v -n" <../stdin
+	)
+'
+
 for opts in '--stdin -z' '-z --stdin'
 do
 	test_expect_success "$opts from subdirectory" '
@@ -648,5 +688,23 @@ do
 	'
 done
 
+test_expect_success PIPE 'streaming support for --stdin' '
+	mkfifo in out &&
+	(git check-ignore -n -v --stdin <in >out &) &&
+
+	# We cannot just "echo >in" because check-ignore would get EOF
+	# after echo exited; instead we open the descriptor in our
+	# shell, and then echo to the fd. We make sure to close it at
+	# the end, so that the subprocess does get EOF and dies
+	# properly.
+	exec 9>in &&
+	test_when_finished "exec 9>&-" &&
+	echo >&9 one &&
+	read response <out &&
+	echo "$response" | grep "^\.gitignore:1:one	one" &&
+	echo >&9 two &&
+	read response <out &&
+	echo "$response" | grep "^::	two"
+'
 
 test_done