Merge branch 'jc/apply-beyond-symlink'

"git apply" was not very careful about reading from, removing,
updating and creating paths outside the working tree (under
--index/--cached) or the current directory (when used as a
replacement for GNU patch).

* jc/apply-beyond-symlink:
  apply: do not touch a file beyond a symbolic link
  apply: do not read from beyond a symbolic link
  apply: do not read from the filesystem under --index
  apply: reject input that touches outside the working area

1 files changed, 141 insertions, 1 deletions

diff --git a/builtin/apply.c b/builtin/apply.c
index 955cc79..525db4a 100644
--- a/builtin/apply.c
+++ b/builtin/apply.c
@@ -51,6 +51,7 @@ static int apply_verbosely;
 static int allow_overlap;
 static int no_add;
 static int threeway;
+static int unsafe_paths;
 static const char *fake_ancestor;
 static int line_termination = '\n';
 static unsigned int p_context = UINT_MAX;
@@ -3221,7 +3222,7 @@ static int load_patch_target(struct strbuf *buf,
 			     const char *name,
 			     unsigned expected_mode)
 {
-	if (cached) {
+	if (cached || check_index) {
 		if (read_file_or_gitlink(ce, buf))
 			return error(_("read of %s failed"), name);
 	} else if (name) {
@@ -3230,6 +3231,8 @@ static int load_patch_target(struct strbuf *buf,
 				return read_file_or_gitlink(ce, buf);
 			else
 				return SUBMODULE_PATCH_WITHOUT_INDEX;
+		} else if (has_symlink_leading_path(name, strlen(name))) {
+			return error(_("reading from '%s' beyond a symbolic link"), name);
 		} else {
 			if (read_old_data(st, name, buf))
 				return error(_("read of %s failed"), name);
@@ -3570,6 +3573,121 @@ static int check_to_create(const char *new_name, int ok_if_exists)
 }
 
 /*
+ * We need to keep track of how symlinks in the preimage are
+ * manipulated by the patches.  A patch to add a/b/c where a/b
+ * is a symlink should not be allowed to affect the directory
+ * the symlink points at, but if the same patch removes a/b,
+ * it is perfectly fine, as the patch removes a/b to make room
+ * to create a directory a/b so that a/b/c can be created.
+ */
+static struct string_list symlink_changes;
+#define SYMLINK_GOES_AWAY 01
+#define SYMLINK_IN_RESULT 02
+
+static uintptr_t register_symlink_changes(const char *path, uintptr_t what)
+{
+	struct string_list_item *ent;
+
+	ent = string_list_lookup(&symlink_changes, path);
+	if (!ent) {
+		ent = string_list_insert(&symlink_changes, path);
+		ent->util = (void *)0;
+	}
+	ent->util = (void *)(what | ((uintptr_t)ent->util));
+	return (uintptr_t)ent->util;
+}
+
+static uintptr_t check_symlink_changes(const char *path)
+{
+	struct string_list_item *ent;
+
+	ent = string_list_lookup(&symlink_changes, path);
+	if (!ent)
+		return 0;
+	return (uintptr_t)ent->util;
+}
+
+static void prepare_symlink_changes(struct patch *patch)
+{
+	for ( ; patch; patch = patch->next) {
+		if ((patch->old_name && S_ISLNK(patch->old_mode)) &&
+		    (patch->is_rename || patch->is_delete))
+			/* the symlink at patch->old_name is removed */
+			register_symlink_changes(patch->old_name, SYMLINK_GOES_AWAY);
+
+		if (patch->new_name && S_ISLNK(patch->new_mode))
+			/* the symlink at patch->new_name is created or remains */
+			register_symlink_changes(patch->new_name, SYMLINK_IN_RESULT);
+	}
+}
+
+static int path_is_beyond_symlink_1(struct strbuf *name)
+{
+	do {
+		unsigned int change;
+
+		while (--name->len && name->buf[name->len] != '/')
+			; /* scan backwards */
+		if (!name->len)
+			break;
+		name->buf[name->len] = '\0';
+		change = check_symlink_changes(name->buf);
+		if (change & SYMLINK_IN_RESULT)
+			return 1;
+		if (change & SYMLINK_GOES_AWAY)
+			/*
+			 * This cannot be "return 0", because we may
+			 * see a new one created at a higher level.
+			 */
+			continue;
+
+		/* otherwise, check the preimage */
+		if (check_index) {
+			struct cache_entry *ce;
+
+			ce = cache_file_exists(name->buf, name->len, ignore_case);
+			if (ce && S_ISLNK(ce->ce_mode))
+				return 1;
+		} else {
+			struct stat st;
+			if (!lstat(name->buf, &st) && S_ISLNK(st.st_mode))
+				return 1;
+		}
+	} while (1);
+	return 0;
+}
+
+static int path_is_beyond_symlink(const char *name_)
+{
+	int ret;
+	struct strbuf name = STRBUF_INIT;
+
+	assert(*name_ != '\0');
+	strbuf_addstr(&name, name_);
+	ret = path_is_beyond_symlink_1(&name);
+	strbuf_release(&name);
+
+	return ret;
+}
+
+static void die_on_unsafe_path(struct patch *patch)
+{
+	const char *old_name = NULL;
+	const char *new_name = NULL;
+	if (patch->is_delete)
+		old_name = patch->old_name;
+	else if (!patch->is_new && !patch->is_copy)
+		old_name = patch->old_name;
+	if (!patch->is_delete)
+		new_name = patch->new_name;
+
+	if (old_name && !verify_path(old_name))
+		die(_("invalid path '%s'"), old_name);
+	if (new_name && !verify_path(new_name))
+		die(_("invalid path '%s'"), new_name);
+}
+
+/*
  * Check and apply the patch in-core; leave the result in patch->result
  * for the caller to write it out to the final destination.
  */
@@ -3656,6 +3774,22 @@ static int check_patch(struct patch *patch)
 		}
 	}
 
+	if (!unsafe_paths)
+		die_on_unsafe_path(patch);
+
+	/*
+	 * An attempt to read from or delete a path that is beyond a
+	 * symbolic link will be prevented by load_patch_target() that
+	 * is called at the beginning of apply_data() so we do not
+	 * have to worry about a patch marked with "is_delete" bit
+	 * here.  We however need to make sure that the patch result
+	 * is not deposited to a path that is beyond a symbolic link
+	 * here.
+	 */
+	if (!patch->is_delete && path_is_beyond_symlink(patch->new_name))
+		return error(_("affected file '%s' is beyond a symbolic link"),
+			     patch->new_name);
+
 	if (apply_data(patch, &st, ce) < 0)
 		return error(_("%s: patch does not apply"), name);
 	patch->rejected = 0;
@@ -3666,6 +3800,7 @@ static int check_patch_list(struct patch *patch)
 {
 	int err = 0;
 
+	prepare_symlink_changes(patch);
 	prepare_fn_table(patch);
 	while (patch) {
 		if (apply_verbosely)
@@ -4404,6 +4539,8 @@ int cmd_apply(int argc, const char **argv, const char *prefix_)
 			N_("make sure the patch is applicable to the current index")),
 		OPT_BOOL(0, "cached", &cached,
 			N_("apply a patch without touching the working tree")),
+		OPT_BOOL(0, "unsafe-paths", &unsafe_paths,
+			N_("accept a patch that touches outside the working area")),
 		OPT_BOOL(0, "apply", &force_apply,
 			N_("also apply the patch (use with --stat/--summary/--check)")),
 		OPT_BOOL('3', "3way", &threeway,
@@ -4476,6 +4613,9 @@ int cmd_apply(int argc, const char **argv, const char *prefix_)
 			die(_("--cached outside a repository"));
 		check_index = 1;
 	}
+	if (check_index)
+		unsafe_paths = 0;
+
 	for (i = 0; i < argc; i++) {
 		const char *arg = argv[i];
 		int fd;