sparse-checkout: match some 'clean' behavior

The 'git sparse-checkout clean' subcommand is somewhat similar to 'git
clean' in that it will delete files that should not be in the worktree.
The big difference is that it focuses on the directories that should not
be in the worktree due to cone-mode sparse-checkout. It also does not
discriminate in the kinds of files and focuses on deleting entire
directories.

However, there are some restrictions that would be good to bring over
from 'git clean', specifically how it refuses to do anything without the
'-f'/'--force' or '-n'/'--dry-run' arguments. The 'clean.requireForce'
config can be set to 'false' to imply '--force'.

Add this behavior to avoid accidental deletion of files that cannot be
recovered from Git.

Signed-off-by: Derrick Stolee <[email protected]>

Author: derrickstolee

Documentation/git-sparse-checkout.adoc

@@ -119,6 +119,15 @@ all sparsity paths.
 	This command can be used to be sure the sparse index works
 	efficiently, though it does not require enabling the sparse index
   feature via the `index.sparse=true` configuration.
++
+To prevent accidental deletion of worktree files, the `clean` subcommand
+will not delete any files without the `-f` or `--force` option, unless
+the `clean.requireForce` config option is set to `false`.
++
+The `--dry-run` option will list the directories that would be removed
+without deleting them. Running in this mode can be helpful to predict the
+behavior of the clean comand or to determine which kinds of files are left
+in the sparse directories.
 
 'disable'::
 	Disable the `core.sparseCheckout` config setting, and restore the

builtin/sparse-checkout.c

@@ -940,6 +940,7 @@ static char const * const builtin_sparse_checkout_clean_usage[] = {
 };
 
 static const char *msg_remove = N_("Removing %s\n");
+static const char *msg_would_remove = N_("Would remove %s\n");
 
 static int sparse_checkout_clean(int argc, const char **argv,
 				   const char *prefix,
@@ -948,8 +949,12 @@ static int sparse_checkout_clean(int argc, const char **argv,
 	struct strbuf full_path = STRBUF_INIT;
 	const char *msg = msg_remove;
 	size_t worktree_len;
+	int force = 0, dry_run = 0;
+	int require_force = 1;
 
 	struct option builtin_sparse_checkout_clean_options[] = {
+		OPT__DRY_RUN(&dry_run, N_("dry run")),
+		OPT__FORCE(&force, N_("force"), PARSE_OPT_NOCOMPLETE),
 		OPT_END(),
 	};
 
@@ -963,6 +968,13 @@ static int sparse_checkout_clean(int argc, const char **argv,
 			     builtin_sparse_checkout_clean_options,
 			     builtin_sparse_checkout_clean_usage, 0);
 
+	repo_config_get_bool(repo, "clean.requireforce", &require_force);
+	if (require_force && !force && !dry_run)
+		die(_("for safety, refusing to clean without one of --force or --dry-run"));
+
+	if (dry_run)
+		msg = msg_would_remove;
+
 	if (repo_read_index(repo) < 0)
 		die(_("failed to read index"));
 
@@ -986,7 +998,8 @@ static int sparse_checkout_clean(int argc, const char **argv,
 
 		printf(msg, ce->name);
 
-		if (remove_dir_recursively(&full_path, 0))
+		if (dry_run <= 0 &&
+		    remove_dir_recursively(&full_path, 0))
 			warning_errno(_("failed to remove '%s'"), ce->name);
 	}
 

t/t1091-sparse-checkout-builtin.sh

@@ -1061,12 +1061,29 @@ test_expect_success 'clean' '
 	touch repo/deep/deeper2/file &&
 	touch repo/folder1/file &&
 
+	test_must_fail git -C repo sparse-checkout clean 2>err &&
+	grep "refusing to clean" err &&
+
+	git -C repo config clean.requireForce true &&
+	test_must_fail git -C repo sparse-checkout clean 2>err &&
+	grep "refusing to clean" err &&
+
+	cat >expect <<-\EOF &&
+	Would remove deep/deeper2/
+	Would remove folder1/
+	EOF
+
+	git -C repo sparse-checkout clean --dry-run >out &&
+	test_cmp expect out &&
+	test_path_exists repo/deep/deeper2 &&
+	test_path_exists repo/folder1 &&
+
 	cat >expect <<-\EOF &&
 	Removing deep/deeper2/
 	Removing folder1/
 	EOF
 
-	git -C repo sparse-checkout clean >out &&
+	git -C repo sparse-checkout clean -f >out &&
 	test_cmp expect out &&
 
 	test_path_is_missing repo/deep/deeper2 &&
@@ -1082,16 +1099,61 @@ test_expect_success 'clean with staged sparse change' '
 
 	git -C repo add --sparse folder1/file &&
 
+	cat >expect <<-\EOF &&
+	Would remove deep/deeper2/
+	EOF
+
+	git -C repo sparse-checkout clean --dry-run >out &&
+	test_cmp expect out &&
+	test_path_exists repo/deep/deeper2 &&
+	test_path_exists repo/folder1 &&
+	test_path_exists repo/folder2 &&
+
 	# deletes deep/deeper2/ but leaves folder1/ and folder2/
 	cat >expect <<-\EOF &&
 	Removing deep/deeper2/
 	EOF
 
+	# The previous test case checked the -f option, so
+	# test the config option in this one.
+	git -C repo config clean.requireForce false &&
 	git -C repo sparse-checkout clean >out &&
 	test_cmp expect out &&
 
 	test_path_is_missing repo/deep/deeper2 &&
-	test_path_exists repo/folder1
+	test_path_exists repo/folder1 &&
+	test_path_exists repo/folder2
+'
+
+test_expect_success 'clean with merge conflict status' '
+	git clone repo clean-merge &&
+
+	echo dirty >clean-merge/deep/deeper2/a &&
+	touch clean-merge/folder2/extra &&
+
+	cat >input <<-EOF &&
+	0 $ZERO_OID	folder1/a
+	100644 $(git -C clean-merge rev-parse HEAD:folder1/a) 1	folder1/a
+	EOF
+	git -C clean-merge update-index --index-info <input &&
+
+	git -C clean-merge sparse-checkout set deep/deeper1 &&
+
+	test_must_fail git -C clean-merge sparse-checkout clean -f 2>err &&
+	grep "failed to convert index to a sparse index" err &&
+
+	mkdir -p clean-merge/folder1/ &&
+	echo merged >clean-merge/folder1/a &&
+	git -C clean-merge add --sparse folder1/a &&
+
+	# deletes folder2/ but leaves staged change in folder1
+	# and dirty change in deep/deeper2/
+	cat >expect <<-\EOF &&
+	Removing folder2/
+	EOF
+
+	git -C clean-merge sparse-checkout clean -f >out &&
+	test_cmp expect out
 '
 
 test_done