git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com>
To: git@vger.kernel.org
Cc: Johannes Schindelin <johannes.schindelin@gmx.de>,
	Johannes Schindelin <johannes.schindelin@gmx.de>
Subject: [PATCH 6/8] hooks(clone protections): special-case current Git LFS hooks
Date: Fri, 17 May 2024 23:15:54 +0000	[thread overview]
Message-ID: <98465797e72cf039ace4138ab1e03e4fc7465ea2.1715987756.git.gitgitgadget@gmail.com> (raw)
In-Reply-To: <pull.1732.git.1715987756.gitgitgadget@gmail.com>

From: Johannes Schindelin <johannes.schindelin@gmx.de>

A notable regression in v2.45.1 and friends (all the way down to
v2.39.4) has been that Git LFS-enabled clones error out with a message
indicating that the `post-checkout` hook has been tampered with while
cloning, and as a safety measure it is not executed.

A generic fix for benign third-party applications wishing to write hooks
during clone operations has been implemented in the parent of this
commit: said applications are expected to add `safe.hook.sha256` values
to a protected config.

However, the current version of Git LFS, v3.5.1, cannot be adapted
retroactively; Therefore, let's just hard-code the SHA-256 values for
this version. That way, Git LFS usage will no longer be broken, and the
next Git LFS version can be taught to add those `safe.hook.sha256`
entries.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 hook.c          | 11 +++++++++++
 t/t1800-hook.sh | 20 ++++++++++++++++++++
 2 files changed, 31 insertions(+)

diff --git a/hook.c b/hook.c
index a2479738451..f810ee133be 100644
--- a/hook.c
+++ b/hook.c
@@ -75,6 +75,17 @@ static int is_hook_safe_during_clone(const char *name, const char *path, char *s
 	if (get_sha256_of_file_contents(path, sha256) < 0)
 		return 0;
 
+	/* Hard-code known-safe values for Git LFS v3.4.0..v3.5.1 */
+	if ((!strcmp("pre-push", name) &&
+	     !strcmp(sha256, "df5417b2daa3aa144c19681d1e997df7ebfe144fb7e3e05138bd80ae998008e4")) ||
+	    (!strcmp("post-checkout", name) &&
+	     !strcmp(sha256, "791471b4ff472aab844a4fceaa48bbb0a12193616f971e8e940625498b4938a6")) ||
+	    (!strcmp("post-commit", name) &&
+	     !strcmp(sha256, "21e961572bb3f43a5f2fbafc1cc764d86046cc2e5f0bbecebfe9684a0b73b664")) ||
+	    (!strcmp("post-merge", name) &&
+	     !strcmp(sha256, "75da0da66a803b4b030ad50801ba57062c6196105eb1d2251590d100edb9390b")))
+		return 1;
+
 	if (!safe_hook_sha256s_initialized) {
 		safe_hook_sha256s_initialized = 1;
 		git_protected_config(safe_hook_cb, &safe_hook_sha256s);
diff --git a/t/t1800-hook.sh b/t/t1800-hook.sh
index 0f74c9154d0..af66999aff3 100755
--- a/t/t1800-hook.sh
+++ b/t/t1800-hook.sh
@@ -192,4 +192,24 @@ test_expect_success '`safe.hook.sha256` and clone protections' '
 	test "called hook" = "$(cat safe-hook/safe-hook.log)"
 '
 
+write_lfs_pre_push_hook () {
+	write_script "$1" <<-\EOF
+	command -v git-lfs >/dev/null 2>&1 || { echo >&2 "\nThis repository is configured for Git LFS but 'git-lfs' was not found on your path. If you no longer wish to use Git LFS, remove this hook by deleting the 'pre-push' file in the hooks directory (set by 'core.hookspath'; usually '.git/hooks').\n"; exit 2; }
+	git lfs pre-push "$@"
+	EOF
+}
+
+test_expect_success 'Git LFS special-handling in clone protections' '
+	git init lfs-hooks &&
+	write_lfs_pre_push_hook lfs-hooks/.git/hooks/pre-push &&
+	write_script git-lfs <<-\EOF &&
+	echo "called $*" >fake-git-lfs.log
+	EOF
+
+	PATH="$PWD:$PATH" GIT_CLONE_PROTECTION_ACTIVE=true \
+		git -C lfs-hooks hook run pre-push &&
+	test_write_lines "called pre-push" >expect &&
+	test_cmp lfs-hooks/fake-git-lfs.log expect
+'
+
 test_done
-- 
gitgitgadget


  parent reply	other threads:[~2024-05-17 23:16 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-05-17 23:15 [PATCH 0/8] Various fixes for v2.45.1 and friends Johannes Schindelin via GitGitGadget
2024-05-17 23:15 ` [PATCH 1/8] hook: plug a new memory leak Johannes Schindelin via GitGitGadget
2024-05-17 23:15 ` [PATCH 2/8] init: use the correct path of the templates directory again Johannes Schindelin via GitGitGadget
2024-05-17 23:15 ` [PATCH 3/8] Revert "core.hooksPath: add some protection while cloning" Johannes Schindelin via GitGitGadget
2024-05-17 23:15 ` [PATCH 4/8] tests: verify that `clone -c core.hooksPath=/dev/null` works again Johannes Schindelin via GitGitGadget
2024-05-18  0:10   ` Junio C Hamano
2024-05-18 18:58     ` Johannes Schindelin
2024-05-17 23:15 ` [PATCH 5/8] hook(clone protections): add escape hatch Johannes Schindelin via GitGitGadget
2024-05-18  0:21   ` Junio C Hamano
2024-05-17 23:15 ` Johannes Schindelin via GitGitGadget [this message]
2024-05-18  0:20   ` [PATCH 6/8] hooks(clone protections): special-case current Git LFS hooks Junio C Hamano
2024-05-17 23:15 ` [PATCH 7/8] hooks(clone protections): simplify templates hooks validation Johannes Schindelin via GitGitGadget
2024-05-17 23:15 ` [PATCH 8/8] Revert "Add a helper function to compare file contents" Johannes Schindelin via GitGitGadget
2024-05-17 23:52 ` [PATCH 0/8] Various fixes for v2.45.1 and friends Junio C Hamano
2024-05-18  0:02   ` Johannes Schindelin
2024-05-18 10:32 ` [PATCH v2 " Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 1/8] hook: plug a new memory leak Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 2/8] init: use the correct path of the templates directory again Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 3/8] Revert "core.hooksPath: add some protection while cloning" Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 4/8] tests: verify that `clone -c core.hooksPath=/dev/null` works again Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 5/8] hook(clone protections): add escape hatch Johannes Schindelin via GitGitGadget
2024-05-18 18:14     ` Jeff King
2024-05-18 18:54       ` Junio C Hamano
2024-05-18 19:35         ` Jeff King
2024-05-18 19:37         ` Johannes Schindelin
2024-05-18 19:32       ` Johannes Schindelin
2024-05-18 19:47         ` Jeff King
2024-05-18 20:06           ` Johannes Schindelin
2024-05-18 21:12             ` Jeff King
2024-05-19  1:15               ` Junio C Hamano
2024-05-20 16:05                 ` Johannes Schindelin
2024-05-20 18:18                   ` Junio C Hamano
2024-05-20 19:38                     ` Johannes Schindelin
2024-05-20 20:07                       ` Junio C Hamano
2024-05-20 21:03                       ` Johannes Schindelin
2024-05-18 10:32   ` [PATCH v2 6/8] hooks(clone protections): special-case current Git LFS hooks Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 7/8] hooks(clone protections): simplify templates hooks validation Johannes Schindelin via GitGitGadget
2024-05-18 10:32   ` [PATCH v2 8/8] Revert "Add a helper function to compare file contents" Johannes Schindelin via GitGitGadget
2024-05-18 17:07   ` [PATCH v2 0/8] Various fixes for v2.45.1 and friends Junio C Hamano
2024-05-18 19:22     ` Johannes Schindelin
2024-05-18 20:13       ` Johannes Schindelin
2024-05-20 20:21   ` [PATCH v3 0/6] " Johannes Schindelin via GitGitGadget
2024-05-20 20:22     ` [PATCH v3 1/6] hook: plug a new memory leak Johannes Schindelin via GitGitGadget
2024-05-20 20:22     ` [PATCH v3 2/6] init: use the correct path of the templates directory again Johannes Schindelin via GitGitGadget
2024-05-20 20:22     ` [PATCH v3 3/6] Revert "core.hooksPath: add some protection while cloning" Johannes Schindelin via GitGitGadget
2024-05-20 20:22     ` [PATCH v3 4/6] tests: verify that `clone -c core.hooksPath=/dev/null` works again Johannes Schindelin via GitGitGadget
2024-05-20 20:22     ` [PATCH v3 5/6] clone: drop the protections where hooks aren't run Johannes Schindelin via GitGitGadget
2024-05-20 20:22     ` [PATCH v3 6/6] Revert "Add a helper function to compare file contents" Johannes Schindelin via GitGitGadget
2024-05-20 23:56     ` [PATCH v3 0/6] Various fixes for v2.45.1 and friends Junio C Hamano
2024-05-21  5:33       ` Junio C Hamano
2024-05-21 18:14         ` Junio C Hamano
2024-05-21 22:33     ` brian m. carlson
2024-05-21 22:40       ` Junio C Hamano
2024-05-21 23:04       ` Junio C Hamano

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=98465797e72cf039ace4138ab1e03e4fc7465ea2.1715987756.git.gitgitgadget@gmail.com \
    --to=gitgitgadget@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=johannes.schindelin@gmx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).