All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Han-Wen Nienhuys via GitGitGadget" <gitgitgadget@gmail.com>
To: git@vger.kernel.org
Cc: Han-Wen Nienhuys <hanwenn@gmail.com>,
	Han-Wen Nienhuys <hanwen@google.com>
Subject: [PATCH 5/7] reftable: ensure that obj_id_len is >= 2 on writing
Date: Wed, 12 Jan 2022 18:07:46 +0000	[thread overview]
Message-ID: <3a72aba447c1405595922bc64b7f7a1a873a033a.1642010868.git.gitgitgadget@gmail.com> (raw)
In-Reply-To: <pull.1185.git.git.1642010868.gitgitgadget@gmail.com>

From: Han-Wen Nienhuys <hanwen@google.com>

When writing the same hash many times, we might decide to use a
length-1 object ID prefix for the ObjectID => ref table, which is out
of spec.

Signed-off-by: Han-Wen Nienhuys <hanwen@google.com>
---
 reftable/readwrite_test.c | 37 +++++++++++++++++++++++++++++++++++++
 reftable/writer.c         |  4 +++-
 2 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/reftable/readwrite_test.c b/reftable/readwrite_test.c
index a315c8992e8..b4371b75724 100644
--- a/reftable/readwrite_test.c
+++ b/reftable/readwrite_test.c
@@ -602,6 +602,42 @@ static void test_write_empty_table(void)
 	strbuf_release(&buf);
 }
 
+static void test_write_object_id_min_length(void)
+{
+	struct reftable_write_options opts = {
+		.block_size = 75,
+	};
+	struct strbuf buf = STRBUF_INIT;
+	struct reftable_writer *w =
+		reftable_new_writer(&strbuf_add_void, &buf, &opts);
+	uint8_t hash[GIT_SHA1_RAWSZ] = {42};
+	struct reftable_ref_record ref = {
+		.update_index = 1,
+		.value_type = REFTABLE_REF_VAL1,
+		.value.val1 = hash,
+	};
+	int err;
+	int i;
+
+	reftable_writer_set_limits(w, 1, 1);
+
+	/* Write the same hash in many refs. If there is only 1 hash, the
+	 * disambiguating prefix is length 0 */
+	for (i = 0; i < 256; i++) {
+		char name[256];
+		snprintf(name, sizeof(name), "ref%05d", i);
+		ref.refname = name;
+		err = reftable_writer_add_ref(w, &ref);
+		EXPECT_ERR(err);
+	}
+
+	err = reftable_writer_close(w);
+	EXPECT_ERR(err);
+	EXPECT(writer_stats(w)->object_id_len == 2);
+	reftable_writer_free(w);
+	strbuf_release(&buf);
+}
+
 static void test_write_empty_key(void)
 {
 	struct reftable_write_options opts = { 0 };
@@ -707,5 +743,6 @@ int readwrite_test_main(int argc, const char *argv[])
 	RUN_TEST(test_write_empty_key);
 	RUN_TEST(test_write_empty_table);
 	RUN_TEST(test_log_overflow);
+	RUN_TEST(test_write_object_id_min_length);
 	return 0;
 }
diff --git a/reftable/writer.c b/reftable/writer.c
index e3c042b9d84..f94af531351 100644
--- a/reftable/writer.c
+++ b/reftable/writer.c
@@ -508,7 +508,9 @@ static void object_record_free(void *void_arg, void *key)
 static int writer_dump_object_index(struct reftable_writer *w)
 {
 	struct write_record_arg closure = { .w = w };
-	struct common_prefix_arg common = { NULL };
+	struct common_prefix_arg common = {
+		.max = 1,		/* obj_id_len should be >= 2. */
+	};
 	if (w->obj_index_tree) {
 		infix_walk(w->obj_index_tree, &update_common, &common);
 	}
-- 
gitgitgadget


  parent reply	other threads:[~2022-01-12 18:08 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-12 18:07 [PATCH 0/7] reftable: avoid reading and writing empty keys Han-Wen Nienhuys via GitGitGadget
2022-01-12 18:07 ` [PATCH 1/7] Documentation: object_id_len goes up to 31 Han-Wen Nienhuys via GitGitGadget
2022-01-12 18:07 ` [PATCH 2/7] reftable: reject 0 object_id_len Han-Wen Nienhuys via GitGitGadget
2022-01-12 18:07 ` [PATCH 3/7] reftable: add a test that verifies that writing empty keys fails Han-Wen Nienhuys via GitGitGadget
2022-01-12 18:07 ` [PATCH 4/7] reftable: avoid writing empty keys at the block layer Han-Wen Nienhuys via GitGitGadget
2022-01-14  1:26   ` Junio C Hamano
2022-01-17 13:10     ` Han-Wen Nienhuys
2022-01-17 19:11       ` Junio C Hamano
2022-01-12 18:07 ` Han-Wen Nienhuys via GitGitGadget [this message]
2022-01-12 18:07 ` [PATCH 6/7] reftable: add test for length of disambiguating prefix Han-Wen Nienhuys via GitGitGadget
2022-01-12 18:07 ` [PATCH 7/7] reftable: rename writer_stats to reftable_writer_stats Han-Wen Nienhuys via GitGitGadget
2022-02-17 13:55 ` [PATCH v2 0/7] reftable: avoid reading and writing empty keys Han-Wen Nienhuys via GitGitGadget
2022-02-17 13:55   ` [PATCH v2 1/7] Documentation: object_id_len goes up to 31 Han-Wen Nienhuys via GitGitGadget
2022-02-17 13:55   ` [PATCH v2 2/7] reftable: reject 0 object_id_len Han-Wen Nienhuys via GitGitGadget
2022-02-18  0:32     ` Junio C Hamano
2022-02-17 13:55   ` [PATCH v2 3/7] reftable: add a test that verifies that writing empty keys fails Han-Wen Nienhuys via GitGitGadget
2022-02-17 13:55   ` [PATCH v2 4/7] reftable: avoid writing empty keys at the block layer Han-Wen Nienhuys via GitGitGadget
2022-02-17 23:55     ` Junio C Hamano
2022-02-21 14:32       ` Han-Wen Nienhuys
2022-02-17 13:55   ` [PATCH v2 5/7] reftable: ensure that obj_id_len is >= 2 on writing Han-Wen Nienhuys via GitGitGadget
2022-02-18  0:01     ` Junio C Hamano
2022-02-17 13:55   ` [PATCH v2 6/7] reftable: add test for length of disambiguating prefix Han-Wen Nienhuys via GitGitGadget
2022-02-17 13:55   ` [PATCH v2 7/7] reftable: rename writer_stats to reftable_writer_stats Han-Wen Nienhuys via GitGitGadget
2022-02-18  0:02   ` [PATCH v2 0/7] reftable: avoid reading and writing empty keys Junio C Hamano
2022-02-21 18:46   ` [PATCH v3 " Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 1/7] Documentation: object_id_len goes up to 31 Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 2/7] reftable: reject 0 object_id_len Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 3/7] reftable: add a test that verifies that writing empty keys fails Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 4/7] reftable: avoid writing empty keys at the block layer Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 5/7] reftable: ensure that obj_id_len is >= 2 on writing Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 6/7] reftable: add test for length of disambiguating prefix Han-Wen Nienhuys via GitGitGadget
2022-02-21 18:46     ` [PATCH v3 7/7] reftable: rename writer_stats to reftable_writer_stats Han-Wen Nienhuys via GitGitGadget
2022-02-23 21:37     ` [PATCH v3 0/7] reftable: avoid reading and writing empty keys 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=3a72aba447c1405595922bc64b7f7a1a873a033a.1642010868.git.gitgitgadget@gmail.com \
    --to=gitgitgadget@gmail.com \
    --cc=git@vger.kernel.org \
    --cc=hanwen@google.com \
    --cc=hanwenn@gmail.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.