From 33ad990ce974be50abdc25427b0f365699d83a34 Mon Sep 17 00:00:00 2001
From: Kyle Evans <kevans@FreeBSD.org>
Date: Thu, 27 Jan 2022 12:02:17 -0600
Subject: [PATCH] cp: fix -R with links

The traversal was previously not properly honoring -H/-L/-P.  Notably,
we should not have been resolving symlinks encountered during traversal
when either -H or -P are specified.

Sponsored by:	Klara, Inc.
Reviewed by:	bapt
Differential Revision:	https://reviews.freebsd.org/D34063
---
 bin/cp/cp.c             | 23 ++++++++++++---
 bin/cp/tests/cp_test.sh | 64 +++++++++++++++++++++++++++++++++++++++++
 2 files changed, 83 insertions(+), 4 deletions(-)

diff --git a/bin/cp/cp.c b/bin/cp/cp.c
index 4328b89cfd7a..7156c7f10742 100644
--- a/bin/cp/cp.c
+++ b/bin/cp/cp.c
@@ -87,7 +87,7 @@ static char emptystring[] = "";
 PATH_T to = { to.p_path, emptystring, "" };
 
 int fflag, iflag, lflag, nflag, pflag, sflag, vflag;
-static int Rflag, rflag;
+static int Hflag, Lflag, Rflag, rflag;
 volatile sig_atomic_t info;
 
 enum op { FILE_TO_FILE, FILE_TO_DIR, DIR_TO_DNE };
@@ -100,11 +100,10 @@ main(int argc, char *argv[])
 {
 	struct stat to_stat, tmp_stat;
 	enum op type;
-	int Hflag, Lflag, ch, fts_options, r, have_trailing_slash;
+	int ch, fts_options, r, have_trailing_slash;
 	char *target;
 
 	fts_options = FTS_NOCHDIR | FTS_PHYSICAL;
-	Hflag = Lflag = 0;
 	while ((ch = getopt(argc, argv, "HLPRafilnprsvx")) != -1)
 		switch (ch) {
 		case 'H':
@@ -271,6 +270,22 @@ main(int argc, char *argv[])
 	    &to_stat)));
 }
 
+/* Does the right thing based on -R + -H/-L/-P */
+static int
+copy_stat(const char *path, struct stat *sb)
+{
+
+	/*
+	 * For -R -H/-P, we need to lstat() instead; copy() cares about the link
+	 * itself rather than the target if we're not following links during the
+	 * traversal.
+	 */
+	if (!Rflag || Lflag)
+		return (stat(path, sb));
+	return (lstat(path, sb));
+}
+
+
 static int
 copy(char *argv[], enum op type, int fts_options, struct stat *root_stat)
 {
@@ -443,7 +458,7 @@ copy(char *argv[], enum op type, int fts_options, struct stat *root_stat)
 		}
 
 		/* Not an error but need to remember it happened. */
-		if (stat(to.p_path, &to_stat) == -1)
+		if (copy_stat(to.p_path, &to_stat) == -1)
 			dne = 1;
 		else {
 			if (to_stat.st_dev == curr->fts_statp->st_dev &&
diff --git a/bin/cp/tests/cp_test.sh b/bin/cp/tests/cp_test.sh
index 3220b45b4438..36d64900f4bb 100755
--- a/bin/cp/tests/cp_test.sh
+++ b/bin/cp/tests/cp_test.sh
@@ -43,6 +43,21 @@ basic_body()
 	check_size baz 4
 }
 
+atf_test_case basic_symlink
+basic_symlink_body()
+{
+	echo "foo" > bar
+	ln -s bar baz
+
+	atf_check cp baz foo
+	atf_check test '!' -L foo
+
+	atf_check -e inline:"cp: baz and baz are identical (not copied).\n" \
+	    -s exit:1 cp baz baz
+	atf_check -e inline:"cp: bar and baz are identical (not copied).\n" \
+	    -s exit:1 cp baz bar
+}
+
 atf_test_case chrdev
 chrdev_body()
 {
@@ -130,12 +145,61 @@ matching_srctgt_nonexistent_body()
 	atf_check -e not-empty -s not-exit:0 stat foo/dne/foo
 }
 
+recursive_link_setup()
+{
+	extra_cpflag=$1
+
+	mkdir -p foo/bar
+	ln -s bar foo/baz
+
+	mkdir foo-mirror
+	eval "cp -R $extra_cpflag foo foo-mirror"
+}
+
+atf_test_case recursive_link_dflt
+recursive_link_dflt_body()
+{
+	recursive_link_setup
+
+	# -P is the default, so this should work and preserve the link.
+	atf_check cp -R foo foo-mirror
+	atf_check test -L foo-mirror/foo/baz
+}
+
+atf_test_case recursive_link_Hflag
+recursive_link_Hflag_body()
+{
+	recursive_link_setup
+
+	# -H will not follow either, so this should also work and preserve the
+	# link.
+	atf_check cp -RH foo foo-mirror
+	atf_check test -L foo-mirror/foo/baz
+}
+
+atf_test_case recursive_link_Lflag
+recursive_link_Lflag_body()
+{
+	recursive_link_setup -L
+
+	# -L will work, but foo/baz ends up expanded to a directory.
+	atf_check test -d foo-mirror/foo/baz -a \
+	    '(' ! -L foo-mirror/foo/baz ')'
+	atf_check cp -RL foo foo-mirror
+	atf_check test -d foo-mirror/foo/baz -a \
+	    '(' ! -L foo-mirror/foo/baz ')'
+}
+
 atf_init_test_cases()
 {
 	atf_add_test_case basic
+	atf_add_test_case basic_symlink
 	atf_add_test_case chrdev
 	atf_add_test_case matching_srctgt
 	atf_add_test_case matching_srctgt_contained
 	atf_add_test_case matching_srctgt_link
 	atf_add_test_case matching_srctgt_nonexistent
+	atf_add_test_case recursive_link_dflt
+	atf_add_test_case recursive_link_Hflag
+	atf_add_test_case recursive_link_Lflag
 }