git/list[1] front-page[2] threads[3] people[4] search[5] about
 

[PATCH] builtin-mv: minimum fix to avoid losing files

From
Junio C Hamano <gitster@pobox.com>
Date
Feb 4, 2008, 08:09 UTC
Message-ID
<7vhcgpkuqz.fsf@gitster.siamese.dyndns.org>
In-Reply-To
<7v3as9mce7.fsf@gitster.siamese.dyndns.org>

An incorrect command "git mv subdir /outer/space" threw the subdirectory to outside of the repository and then noticed that /outer/space/subdir/ would be outside of the repository. The error checking is backwards.

This fixes the issue by being careful about use of the return value of get_pathspec(). Since the implementation already has handcrafted loop to munge each path on the command line, we use prefix_path() instead.

Signed-off-by: Junio C Hamano <gitster@pobox.com>
---
 * This minimally fixes the issue and applies on top of the
   "setup: sanitize absolute and funny paths in get_pathspec()"
   patch I showed during the rc freeze.
   Dscho CC'ed as he owns the largest number of lines in this
   source file.
 builtin-mv.c  |    6 +++++-
 t/t7001-mv.sh |   38 ++++++++++++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+), 1 deletions(-)
diff --git a/builtin-mv.c b/builtin-mv.c
index 94f6dd2..68aa2a6 100644
--- a/builtin-mv.c
+++ b/builtin-mv.c
@@ -19,6 +19,7 @@ static const char **copy_pathspec(const char *prefix, const char **pathspec,
 				  int count, int base_name)
 {
 	int i;
+	int len = prefix ? strlen(prefix) : 0;
 	const char **result = xmalloc((count + 1) * sizeof(const char *));
 	memcpy(result, pathspec, count * sizeof(const char *));
 	result[count] = NULL;
@@ -32,8 +33,11 @@ static const char **copy_pathspec(const char *prefix, const char **pathspec,
 			if (last_slash)
 				result[i] = last_slash + 1;
 		}
+		result[i] = prefix_path(prefix, len, result[i]);
+		if (!result[i])
+			exit(1); /* error already given */
 	}
-	return get_pathspec(prefix, result);
+	return result;
 }
 
 static void show_list(const char *label, struct path_list *list)
diff --git a/t/t7001-mv.sh b/t/t7001-mv.sh
index b1243b4..fa382c5 100755
--- a/t/t7001-mv.sh
+++ b/t/t7001-mv.sh
@@ -118,4 +118,42 @@ test_expect_success "Sergey Vlasov's test case" '
 	git mv ab a
 '
 
+test_expect_success 'absolute pathname' '(
+
+	rm -fr mine &&
+	mkdir mine &&
+	cd mine &&
+	test_create_repo one &&
+	cd one &&
+	mkdir sub &&
+	>sub/file &&
+	git add sub/file &&
+
+	git mv sub "$(pwd)/in" &&
+	! test -d sub &&
+	test -d in &&
+	git ls-files --error-unmatch in/file
+
+
+)'
+
+test_expect_success 'absolute pathname outside should fail' '(
+
+	rm -fr mine &&
+	mkdir mine &&
+	cd mine &&
+	out=$(pwd) &&
+	test_create_repo one &&
+	cd one &&
+	mkdir sub &&
+	>sub/file &&
+	git add sub/file &&
+
+	! git mv sub "$out/out" &&
+	test -d sub &&
+	! test -d ../in &&
+	git ls-files --error-unmatch sub/file
+
+)'
+
 test_done
-- 
1.5.4.18.gd0b8
Previous: Junio C HamanoNext: Johannes Schindelin
Message 2 of 6 in “fix misuse of prefix_path()”
  1. fix misuse of prefix_path()Junio C Hamano, Feb 4, 2008
  2. builtin-mv: minimum fix to avoid losing filesJunio C Hamano, Feb 4, 2008
  3. Johannes SchindelinFeb 4, 2008
  4. Johannes SixtFeb 4, 2008
  5. Fix misuse of prefix_path()Johannes Sixt, Feb 5, 2008
  6. Junio C HamanoFeb 5, 2008

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.