From cc77a004cea178f3eae02f45e55278cd74a5e205 Mon Sep 17 00:00:00 2001
From: Elijah Newren <newren@gmail.com>
Date: Mon, 29 Jul 2024 15:41:02 -0700
Subject: [PATCH] fast-import: avoid making replace refs point to themselves

If someone replaces a commit with a modified version, then builds on
that commit, and then later decides to rewrite history in a format like

    git fast-export --all | CMD_TO_TWEAK_THE_STREAM | git fast-import

and CMD_TO_TWEAK_THE_STREAM undoes the modifications that the
replacement did, then at the end you'd get a replace ref that points to
itself.  For example:

    $ git show-ref | grep replace
    fb92ebc654641b310e7d0360d0a5a49316fd7264 refs/replace/fb92ebc654641b310e7d0360d0a5a49316fd7264

Git commands which pay attention to replace refs will die with an error
when a self-referencing replace ref is present:

    $ git log
    fatal: replace depth too high for object fb92ebc654641b310e7d0360d0a5a49316fd7264

Avoid such problems by deleting replace refs that will simply end up
pointing to themselves at the end of our writing.  Unless users specify
--quiet, warn them when we delete such a replace ref.

Two notes about this patch:
  * We are not ignoring the problematic update of the replace ref
    (turning it into a no-op), we are replacing the update with a delete.
    The logic here is that if the repository had a value for the replace
    ref before fast-import was run, and the replace ref was explicitly
    named in the fast-import stream, we don't want the replace ref to be
    left with a pre-fast-import value.
  * While loops with more than one element (e.g. refs/replace/A points
    to B, and refs/replace/B points to A) are possible, they seem much
    less plausible.  It is pretty easy to create a sequence of
    git-filter-repo commands that will trigger a self-referencing replace
    ref, but I do not know how to trigger a scenario with a cycle length
    greater than 1.

Signed-off-by: Elijah Newren <newren@gmail.com>
---
 builtin/fast-import.c  | 16 +++++++++++++++-
 t/t9300-fast-import.sh | 28 ++++++++++++++++++++++++++++
 2 files changed, 43 insertions(+), 1 deletion(-)

diff --git a/builtin/fast-import.c b/builtin/fast-import.c
index 76d5c20f141f42..51c8228cb7b0b4 100644
--- a/builtin/fast-import.c
+++ b/builtin/fast-import.c
@@ -179,6 +179,7 @@ static unsigned long branch_load_count;
 static int failure;
 static FILE *pack_edges;
 static unsigned int show_stats = 1;
+static unsigned int quiet;
 static int global_argc;
 static const char **global_argv;
 static const char *global_prefix;
@@ -1602,7 +1603,19 @@ static int update_branch(struct branch *b)
 	struct ref_transaction *transaction;
 	struct object_id old_oid;
 	struct strbuf err = STRBUF_INIT;
-
+	static const char *replace_prefix = "refs/replace/";
+
+	if (starts_with(b->name, replace_prefix) &&
+	    !strcmp(b->name + strlen(replace_prefix),
+		    oid_to_hex(&b->oid))) {
+		if (!quiet)
+			warning("Dropping %s since it would point to "
+				"itself (i.e. to %s)",
+				b->name, oid_to_hex(&b->oid));
+		refs_delete_ref(get_main_ref_store(the_repository),
+				NULL, b->name, NULL, 0);
+		return 0;
+	}
 	if (is_null_oid(&b->oid)) {
 		if (b->delete)
 			refs_delete_ref(get_main_ref_store(the_repository),
@@ -3388,6 +3401,7 @@ static int parse_one_option(const char *option)
 		option_export_pack_edges(option);
 	} else if (!strcmp(option, "quiet")) {
 		show_stats = 0;
+		quiet = 1;
 	} else if (!strcmp(option, "stats")) {
 		show_stats = 1;
 	} else if (!strcmp(option, "allow-unsafe-features")) {
diff --git a/t/t9300-fast-import.sh b/t/t9300-fast-import.sh
index 6224f54d4d26ce..425a261c161692 100755
--- a/t/t9300-fast-import.sh
+++ b/t/t9300-fast-import.sh
@@ -3692,6 +3692,34 @@ test_expect_success ICONV 'X: handling encoding' '
 	git log -1 --format=%B encoding | grep $(printf "\317\200")
 '
 
+test_expect_success 'X: replace ref that becomes useless is removed' '
+	git init -qb main testrepo &&
+	cd testrepo &&
+	(
+		test_commit test &&
+
+		test_commit msg somename content &&
+
+		git mv somename othername &&
+		NEW_TREE=$(git write-tree) &&
+		MSG="$(git log -1 --format=%B HEAD)" &&
+		NEW_COMMIT=$(git commit-tree -p HEAD^1 -m "$MSG" $NEW_TREE) &&
+		git replace main $NEW_COMMIT &&
+
+		echo more >>othername &&
+		git add othername &&
+		git commit -qm more &&
+
+		git fast-export --all >tmp &&
+		sed -e s/othername/somename/ tmp >tmp2 &&
+		git fast-import --force <tmp2 2>msgs &&
+
+		grep "Dropping.*since it would point to itself" msgs &&
+		git show-ref >refs &&
+		! grep refs/replace refs
+	)
+'
+
 ###
 ### series Y (submodules and hash algorithms)
 ###