Merge branch 'bc/use-sha256-by-default-in-3.0'

gitster · gitster · commit a636d395ff0b · 2025-07-21T09:14:25.000-07:00
Prepare to flip the default hash function to SHA-256.

* bc/use-sha256-by-default-in-3.0:
  Enable SHA-256 by default in breaking changes mode
  help: add a build option for default hash
  t5300: choose the built-in hash outside of a repo
  t4042: choose the built-in hash outside of a repo
  t1007: choose the built-in hash outside of a repo
  t: default to compile-time default hash if not set
  setup: use the default algorithm to initialize repo format
  Use legacy hash for legacy formats
  builtin: use default hash when outside a repository
  hash: add a constant for the legacy hash algorithm
  hash: add a constant for the default hash algorithm
diff --git a/builtin/apply.c b/builtin/apply.c
@@ -29,7 +29,7 @@ int cmd_apply(int argc,
 	 * cf. https://lore.kernel.org/git/xmqqcypfcmn4.fsf@gitster.g/
 	 */
 	if (!the_hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	argc = apply_parse_options(argc, argv,
 				   &state, &force_apply, &options,
diff --git a/builtin/diff.c b/builtin/diff.c
@@ -483,7 +483,7 @@ int cmd_diff(int argc,
 	 * configurable via a command line option.
 	 */
 	if (nongit)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	init_diff_ui_defaults();
 	git_config(git_diff_ui_config, NULL);
diff --git a/builtin/hash-object.c b/builtin/hash-object.c
@@ -104,7 +104,7 @@ int cmd_hash_object(int argc,
 		prefix = setup_git_directory_gently(&nongit);
 
 	if (nongit && !the_hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	if (vpath && prefix) {
 		vpath_free = prefix_filename(prefix, vpath);
diff --git a/builtin/index-pack.c b/builtin/index-pack.c
@@ -2035,7 +2035,7 @@ int cmd_index_pack(int argc,
 	 * choice but to guess the object hash.
 	 */
 	if (!the_repository->hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	opts.flags &= ~(WRITE_REV | WRITE_REV_VERIFY);
 	if (rev_index) {
diff --git a/builtin/ls-remote.c b/builtin/ls-remote.c
@@ -112,7 +112,7 @@ int cmd_ls_remote(int argc,
 	 * depending on what object hash the remote uses.
 	 */
 	if (!the_repository->hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	packet_trace_identity("ls-remote");
 
diff --git a/builtin/patch-id.c b/builtin/patch-id.c
@@ -254,7 +254,7 @@ int cmd_patch_id(int argc,
 	 * the code that computes patch IDs to always use SHA1.
 	 */
 	if (!the_hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	generate_id_list(opts ? opts > 1 : config.stable,
 			 opts ? opts == 3 : config.verbatim);
diff --git a/builtin/receive-pack.c b/builtin/receive-pack.c
@@ -2203,7 +2203,7 @@ static struct command *read_head_info(struct packet_reader *reader,
 				use_push_options = 1;
 			hash = parse_feature_value(feature_list, "object-format", &len, NULL);
 			if (!hash) {
-				hash = hash_algos[GIT_HASH_SHA1].name;
+				hash = hash_algos[GIT_HASH_SHA1_LEGACY].name;
 				len = strlen(hash);
 			}
 			if (xstrncmpz(the_hash_algo->name, hash, len))
diff --git a/builtin/shortlog.c b/builtin/shortlog.c
@@ -419,7 +419,7 @@ int cmd_shortlog(int argc,
 	 * git/nongit so that we do not have to do this.
 	 */
 	if (nongit && !the_hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	git_config(git_default_config, NULL);
 	shortlog_init(&log);
diff --git a/builtin/show-index.c b/builtin/show-index.c
@@ -47,7 +47,7 @@ int cmd_show_index(int argc,
 	 *       the index file passed in and use that instead.
 	 */
 	if (!the_hash_algo)
-		repo_set_hash_algo(the_repository, GIT_HASH_SHA1);
+		repo_set_hash_algo(the_repository, GIT_HASH_DEFAULT);
 
 	hashsz = the_hash_algo->rawsz;
 
diff --git a/bundle.c b/bundle.c
@@ -95,7 +95,7 @@ int read_bundle_header_fd(int fd, struct bundle_header *header,
 	 * by an "object-format=" capability, which is being handled in
 	 * `parse_capability()`.
 	 */
-	header->hash_algo = &hash_algos[GIT_HASH_SHA1];
+	header->hash_algo = &hash_algos[GIT_HASH_SHA1_LEGACY];
 
 	/* The bundle header ends with an empty line */
 	while (!strbuf_getwholeline_fd(&buf, fd, '\n') &&
@@ -507,7 +507,7 @@ int create_bundle(struct repository *r, const char *path,
 	 *    SHA1.
 	 * 2. @filter is required because we parsed an object filter.
 	 */
-	if (the_hash_algo != &hash_algos[GIT_HASH_SHA1] || revs.filter.choice)
+	if (the_hash_algo != &hash_algos[GIT_HASH_SHA1_LEGACY] || revs.filter.choice)
 		min_version = 3;
 
 	if (argc > 1) {
diff --git a/connect.c b/connect.c
@@ -251,7 +251,7 @@ static void process_capabilities(struct packet_reader *reader, size_t *linelen)
 			reader->hash_algo = &hash_algos[hash_algo];
 		free(hash_name);
 	} else {
-		reader->hash_algo = &hash_algos[GIT_HASH_SHA1];
+		reader->hash_algo = &hash_algos[GIT_HASH_SHA1_LEGACY];
 	}
 }
 
@@ -500,7 +500,7 @@ static void send_capabilities(int fd_out, struct packet_reader *reader)
 		reader->hash_algo = &hash_algos[hash_algo];
 		packet_write_fmt(fd_out, "object-format=%s", reader->hash_algo->name);
 	} else {
-		reader->hash_algo = &hash_algos[GIT_HASH_SHA1];
+		reader->hash_algo = &hash_algos[GIT_HASH_SHA1_LEGACY];
 	}
 	if (server_feature_v2("promisor-remote", &promisor_remote_info)) {
 		char *reply = promisor_remote_reply(promisor_remote_info);
@@ -665,7 +665,7 @@ int server_supports_hash(const char *desired, int *feature_supported)
 	if (feature_supported)
 		*feature_supported = !!hash;
 	if (!hash) {
-		hash = hash_algos[GIT_HASH_SHA1].name;
+		hash = hash_algos[GIT_HASH_SHA1_LEGACY].name;
 		len = strlen(hash);
 	}
 	while (hash) {
diff --git a/fetch-pack.c b/fetch-pack.c
@@ -1343,7 +1343,7 @@ static void write_fetch_command_and_capabilities(struct strbuf *req_buf,
 			die(_("mismatched algorithms: client %s; server %s"),
 			    the_hash_algo->name, hash_name);
 		packet_buf_write(req_buf, "object-format=%s", the_hash_algo->name);
-	} else if (hash_algo_by_ptr(the_hash_algo) != GIT_HASH_SHA1) {
+	} else if (hash_algo_by_ptr(the_hash_algo) != GIT_HASH_SHA1_LEGACY) {
 		die(_("the server does not support algorithm '%s'"),
 		    the_hash_algo->name);
 	}
diff --git a/hash.h b/hash.h
@@ -175,6 +175,16 @@ static inline void git_SHA256_Clone(git_SHA256_CTX *dst, const git_SHA256_CTX *s
 /* Number of algorithms supported (including unknown). */
 #define GIT_HASH_NALGOS (GIT_HASH_SHA256 + 1)
 
+/* Default hash algorithm if unspecified. */
+#ifdef WITH_BREAKING_CHANGES
+# define GIT_HASH_DEFAULT GIT_HASH_SHA256
+#else
+# define GIT_HASH_DEFAULT GIT_HASH_SHA1
+#endif
+
+/* Legacy hash algorithm. Implied for older data formats which don't specify. */
+#define GIT_HASH_SHA1_LEGACY GIT_HASH_SHA1
+
 /* "sha1", big-endian */
 #define GIT_SHA1_FORMAT_ID 0x73686131
 
diff --git a/help.c b/help.c
@@ -812,6 +812,7 @@ void get_version_info(struct strbuf *buf, int show_build_options)
 		strbuf_addf(buf, "SHA-256: %s\n", SHA256_BACKEND);
 		strbuf_addf(buf, "default-ref-format: %s\n",
 			    ref_storage_format_to_name(REF_STORAGE_FORMAT_DEFAULT));
+		strbuf_addf(buf, "default-hash: %s\n", hash_algos[GIT_HASH_DEFAULT].name);
 	}
 }
 
diff --git a/pkt-line.c b/pkt-line.c
@@ -617,7 +617,7 @@ void packet_reader_init(struct packet_reader *reader, int fd,
 	reader->buffer_size = sizeof(packet_buffer);
 	reader->options = options;
 	reader->me = "git";
-	reader->hash_algo = &hash_algos[GIT_HASH_SHA1];
+	reader->hash_algo = &hash_algos[GIT_HASH_SHA1_LEGACY];
 	strbuf_init(&reader->scratch, 0);
 }
 
diff --git a/remote-curl.c b/remote-curl.c
@@ -285,7 +285,7 @@ static const struct git_hash_algo *detect_hash_algo(struct discovery *heads)
 	 * back to SHA1, which may or may not be correct.
 	 */
 	if (!p)
-		return &hash_algos[GIT_HASH_SHA1];
+		return &hash_algos[GIT_HASH_SHA1_LEGACY];
 
 	algo = hash_algo_by_length((p - heads->buf) / 2);
 	if (algo == GIT_HASH_UNKNOWN)
diff --git a/serve.c b/serve.c
@@ -14,7 +14,7 @@
 
 static int advertise_sid = -1;
 static int advertise_object_info = -1;
-static int client_hash_algo = GIT_HASH_SHA1;
+static int client_hash_algo = GIT_HASH_SHA1_LEGACY;
 
 static int always_advertise(struct repository *r UNUSED,
 			    struct strbuf *value UNUSED)
diff --git a/setup.c b/setup.c
@@ -836,9 +836,12 @@ static void init_repository_format(struct repository_format *format)
 int read_repository_format(struct repository_format *format, const char *path)
 {
 	clear_repository_format(format);
+	format->hash_algo = GIT_HASH_SHA1_LEGACY;
 	git_config_from_file(check_repo_format, path, format);
-	if (format->version == -1)
+	if (format->version == -1) {
 		clear_repository_format(format);
+		format->hash_algo = GIT_HASH_SHA1_LEGACY;
+	}
 	return format->version;
 }
 
@@ -2225,11 +2228,11 @@ void initialize_repository_version(int hash_algo,
 	 * version will get adjusted by git-clone(1) once it has learned about
 	 * the remote repository's format.
 	 */
-	if (hash_algo != GIT_HASH_SHA1 ||
+	if (hash_algo != GIT_HASH_SHA1_LEGACY ||
 	    ref_storage_format != REF_STORAGE_FORMAT_FILES)
 		target_version = GIT_REPO_VERSION_READ;
 
-	if (hash_algo != GIT_HASH_SHA1 && hash_algo != GIT_HASH_UNKNOWN)
+	if (hash_algo != GIT_HASH_SHA1_LEGACY && hash_algo != GIT_HASH_UNKNOWN)
 		git_config_set("extensions.objectformat",
 			       hash_algos[hash_algo].name);
 	else if (reinit)
diff --git a/setup.h b/setup.h
@@ -149,7 +149,7 @@ struct repository_format {
 { \
 	.version = -1, \
 	.is_bare = -1, \
-	.hash_algo = GIT_HASH_SHA1, \
+	.hash_algo = GIT_HASH_DEFAULT, \
 	.ref_storage_format = REF_STORAGE_FORMAT_FILES, \
 	.unknown_extensions = STRING_LIST_INIT_DUP, \
 	.v1_only_extensions = STRING_LIST_INIT_DUP, \
diff --git a/t/t1007-hash-object.sh b/t/t1007-hash-object.sh
@@ -252,9 +252,9 @@ test_expect_success '--literally complains about non-standard types' '
 	test_must_fail git hash-object -t bogus --literally --stdin
 '
 
-test_expect_success '--stdin outside of repository (uses SHA-1)' '
+test_expect_success '--stdin outside of repository (uses default hash)' '
 	nongit git hash-object --stdin <hello >actual &&
-	echo "$(test_oid --hash=sha1 hello)" >expect &&
+	echo "$(test_oid --hash=builtin hello)" >expect &&
 	test_cmp expect actual
 '
 
diff --git a/t/t4042-diff-textconv-caching.sh b/t/t4042-diff-textconv-caching.sh
@@ -120,6 +120,14 @@ test_expect_success 'log notes cache and still use cache for -p' '
 '
 
 test_expect_success 'caching is silently ignored outside repo' '
+	test_oid_cache <<-\EOM &&
+	oid1 sha1:5626abf
+	oid1 sha256:a4ed1f3
+	oid2 sha1:f719efd
+	oid2 sha256:aa9e7dc
+	EOM
+	oid1=$(test_oid --hash=builtin oid1) &&
+	oid2=$(test_oid --hash=builtin oid2) &&
 	mkdir -p non-repo &&
 	echo one >non-repo/one &&
 	echo two >non-repo/two &&
@@ -129,9 +137,9 @@ test_expect_success 'caching is silently ignored outside repo' '
 		   -c diff.test.textconv="tr a-z A-Z <" \
 		   -c diff.test.cachetextconv=true \
 		   diff --no-index one two >actual &&
-	cat >expect <<-\EOF &&
+	cat >expect <<-EOF &&
 	diff --git a/one b/two
-	index 5626abf..f719efd 100644
+	index $oid1..$oid2 100644
 	--- a/one
 	+++ b/two
 	@@ -1 +1 @@
diff --git a/t/t5300-pack-object.sh b/t/t5300-pack-object.sh
@@ -525,7 +525,7 @@ test_expect_success 'index-pack --strict <pack> works in non-repo' '
 	test_path_is_file foo.idx
 '
 
-test_expect_success SHA1 'show-index works OK outside a repository' '
+test_expect_success DEFAULT_HASH_ALGORITHM 'show-index works OK outside a repository' '
 	nongit git show-index <foo.idx
 '
 
@@ -658,7 +658,7 @@ do
 		test_commit -C repo initial &&
 		git -C repo repack -ad &&
 		git -C repo verify-pack "$(pwd)"/repo/.git/objects/pack/*.idx &&
-		if test $hash = sha1
+		if test $hash = $GIT_TEST_BUILTIN_HASH
 		then
 			nongit git verify-pack "$(pwd)"/repo/.git/objects/pack/*.idx
 		else
@@ -676,7 +676,7 @@ do
 		test_commit -C repo initial &&
 		git -C repo repack -ad &&
 		git -C repo index-pack --verify "$(pwd)"/repo/.git/objects/pack/*.pack &&
-		if test $hash = sha1
+		if test $hash = $GIT_TEST_BUILTIN_HASH
 		then
 			nongit git index-pack --verify "$(pwd)"/repo/.git/objects/pack/*.pack
 		else
diff --git a/t/test-lib-functions.sh b/t/test-lib-functions.sh
@@ -1707,7 +1707,7 @@ test_set_hash () {
 
 # Detect the hash algorithm in use.
 test_detect_hash () {
-	case "$GIT_TEST_DEFAULT_HASH" in
+	case "${GIT_TEST_DEFAULT_HASH:-$GIT_TEST_BUILTIN_HASH}" in
 	"sha256")
 	    test_hash_algo=sha256
 	    test_compat_hash_algo=sha1
@@ -1779,6 +1779,9 @@ test_oid () {
 	--hash=compat)
 		algo="$test_compat_hash_algo" &&
 		shift;;
+	--hash=builtin)
+		algo="$GIT_TEST_BUILTIN_HASH" &&
+		shift;;
 	--hash=*)
 		algo="${1#--hash=}" &&
 		shift;;
diff --git a/t/test-lib.sh b/t/test-lib.sh
@@ -536,7 +536,8 @@ export GIT_COMMITTER_EMAIL GIT_COMMITTER_NAME
 export GIT_COMMITTER_DATE GIT_AUTHOR_DATE
 export EDITOR
 
-GIT_DEFAULT_HASH="${GIT_TEST_DEFAULT_HASH:-sha1}"
+GIT_TEST_BUILTIN_HASH=$("$GIT_BUILD_DIR/git" version --build-options | sed -ne 's/^default-hash: //p')
+GIT_DEFAULT_HASH="${GIT_TEST_DEFAULT_HASH:-$GIT_TEST_BUILTIN_HASH}"
 export GIT_DEFAULT_HASH
 GIT_DEFAULT_REF_FORMAT="${GIT_TEST_DEFAULT_REF_FORMAT:-files}"
 export GIT_DEFAULT_REF_FORMAT
@@ -1908,6 +1909,10 @@ test_lazy_prereq SHA1 '
 	esac
 '
 
+test_lazy_prereq DEFAULT_HASH_ALGORITHM '
+	test "$GIT_TEST_BUILTIN_HASH" = "$GIT_DEFAULT_HASH"
+'
+
 test_lazy_prereq DEFAULT_REPO_FORMAT '
 	test_have_prereq SHA1,REFFILES
 '
diff --git a/transport.c b/transport.c
@@ -1243,7 +1243,7 @@ struct transport *transport_get(struct remote *remote, const char *url)
 			ret->smart_options->receivepack = remote->receivepack;
 	}
 
-	ret->hash_algo = &hash_algos[GIT_HASH_SHA1];
+	ret->hash_algo = &hash_algos[GIT_HASH_SHA1_LEGACY];
 
 	return ret;
 }

Original file line number	Diff line number	Diff line change
`@@ -2203,7 +2203,7 @@ static struct command read_head_info(struct packet_reader reader,`
`2203`	`2203`	`use_push_options = 1;`
`2204`	`2204`	`hash = parse_feature_value(feature_list, "object-format", &len, NULL);`
`2205`	`2205`	`if (!hash) {`
`2206`		`- hash = hash_algos[GIT_HASH_SHA1].name;`
	`2206`	`+ hash = hash_algos[GIT_HASH_SHA1_LEGACY].name;`
`2207`	`2207`	`len = strlen(hash);`
`2208`	`2208`	`}`
`2209`	`2209`	`if (xstrncmpz(the_hash_algo->name, hash, len))`