object-file API: split up and simplify check_object_signature()
Split up the check_object_signature() function into that non-streaming version (it accepts an already filled "buf"), and a new stream_object_signature() which will retrieve the object from storage, and hash it on-the-fly. All of the callers of check_object_signature() were effectively calling two different functions, if we go by cyclomatic complexity. I.e. they'd either take the early "if (map)" branch and return early, or not. This has been the case since the "if (map)" condition was added in 090ea12671b (parse_object: avoid putting whole blob in core, 2012-03-07). We can then further simplify the resulting check_object_signature() function since only one caller wanted to pass a non-NULL "buf" and a non-NULL "real_oidp". That "read_loose_object()" codepath used by "git fsck" can instead use hash_object_file() followed by oideq(). Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Ævar Arnfjörð Bjarmason committed
Feb 5, 2022 at 00:48 UTC
0f156dbb04b434d95ce5465e6b07d8869d55e8e0
7 files changed
+38
-28
builtin/fast-export.c
+1
-1
@@ -300,7 +300,7 @@ static void export_blob(const struct object_id *oid)
300
if (!buf)
301
die("could not read blob %s", oid_to_hex(oid));
302
if (check_object_signature(the_repository, oid, buf, size,
303
- type_name(type), NULL) < 0)
303
+ type_name(type)) < 0)
304
die("oid mismatch in blob %s", oid_to_hex(oid));
305
object = parse_object_buffer(the_repository, oid, type,
306
size, buf, &eaten);
builtin/index-pack.c
+1
-1
@@ -1413,7 +1413,7 @@ static void fix_unresolved_deltas(struct hashfile *f)
1413
continue;
1414
1415
if (check_object_signature(the_repository, &d->oid, data, size,
1416
- type_name(type), NULL) < 0)
1416
+ type_name(type)) < 0)
1417
die(_("local object %s is corrupt"), oid_to_hex(&d->oid));
1418
1419
/*
builtin/mktag.c
+2
-3
@@ -61,9 +61,8 @@ static int verify_object_in_tag(struct object_id *tagged_oid, int *tagged_type)
61
type_name(*tagged_type), type_name(type));
62
63
repl = lookup_replace_object(the_repository, tagged_oid);
64
- ret = check_object_signature(the_repository, repl,
65
- buffer, size, type_name(*tagged_type),
66
- NULL);
64
+ ret = check_object_signature(the_repository, repl, buffer, size,
65
+ type_name(*tagged_type));
66
free(buffer);
67
68
return ret;
cache.h
+8
-4
@@ -1322,15 +1322,19 @@ int parse_loose_header(const char *hdr, struct object_info *oi);
1322
/**
1323
* With in-core object data in "buf", rehash it to make sure the
1324
* object name actually matches "oid" to detect object corruption.
1325
- * With "buf" == NULL, try reading the object named with "oid" using
1326
- * the streaming interface and rehash it to do the same.
1325
*
1326
* A negative value indicates an error, usually that the OID is not
1327
* what we expected, but it might also indicate another error.
1328
*/
1329
int check_object_signature(struct repository *r, const struct object_id *oid,
1332
- void *buf, unsigned long size, const char *type,
1333
- struct object_id *real_oidp);
1330
+ void *buf, unsigned long size, const char *type);
1331
+
1332
+/**
1333
+ * A streaming version of check_object_signature().
1334
+ * Try reading the object named with "oid" using
1335
+ * the streaming interface and rehash it to do the same.
1336
+ */
1337
+int stream_object_signature(struct repository *r, const struct object_id *oid);
1338
1339
int finalize_object_file(const char *tmpfile, const char *filename);
1340
object-file.c
+18
-14
@@ -1067,22 +1067,25 @@ int format_object_header(char *str, size_t size, enum object_type type,
1067
}
1068
1069
int check_object_signature(struct repository *r, const struct object_id *oid,
1070
- void *buf, unsigned long size, const char *type,
1071
- struct object_id *real_oidp)
1070
+ void *buf, unsigned long size, const char *type)
1071
{
1073
- struct object_id tmp;
1074
- struct object_id *real_oid = real_oidp ? real_oidp : &tmp;
1072
+ struct object_id real_oid;
1073
+
1074
+ hash_object_file(r->hash_algo, buf, size, type, &real_oid);
1075
+
1076
+ return !oideq(oid, &real_oid) ? -1 : 0;
1077
+}
1078
+
1079
+int stream_object_signature(struct repository *r, const struct object_id *oid)
1080
+{
1081
+ struct object_id real_oid;
1082
+ unsigned long size;
1083
enum object_type obj_type;
1084
struct git_istream *st;
1085
git_hash_ctx c;
1086
char hdr[MAX_HEADER_LEN];
1087
int hdrlen;
1088
1081
- if (buf) {
1082
- hash_object_file(r->hash_algo, buf, size, type, real_oid);
1083
- return !oideq(oid, real_oid) ? -1 : 0;
1084
- }
1085
-
1089
st = open_istream(r, oid, &obj_type, &size, NULL);
1090
if (!st)
1091
return -1;
@@ -1105,9 +1108,9 @@ int check_object_signature(struct repository *r, const struct object_id *oid,
1108
break;
1109
r->hash_algo->update_fn(&c, buf, readlen);
1110
}
1108
- r->hash_algo->final_oid_fn(real_oid, &c);
1111
+ r->hash_algo->final_oid_fn(&real_oid, &c);
1112
close_istream(st);
1110
- return !oideq(oid, real_oid) ? -1 : 0;
1113
+ return !oideq(oid, &real_oid) ? -1 : 0;
1114
}
1115
1116
int git_open_cloexec(const char *name, int flags)
@@ -2611,9 +2614,10 @@ int read_loose_object(const char *path,
2614
git_inflate_end(&stream);
2615
goto out;
2616
}
2614
- if (check_object_signature(the_repository, expected_oid,
2615
- *contents, *size,
2616
- oi->type_name->buf, real_oid) < 0)
2617
+ hash_object_file(the_repository->hash_algo,
2618
+ *contents, *size, oi->type_name->buf,
2619
+ real_oid);
2620
+ if (!oideq(expected_oid, real_oid))
2621
goto out;
2622
}
2623
object.c
+2
-2
@@ -279,7 +279,7 @@ struct object *parse_object(struct repository *r, const struct object_id *oid)
279
if ((obj && obj->type == OBJ_BLOB && repo_has_object_file(r, oid)) ||
280
(!obj && repo_has_object_file(r, oid) &&
281
oid_object_info(r, oid, NULL) == OBJ_BLOB)) {
282
- if (check_object_signature(r, repl, NULL, 0, NULL, NULL) < 0) {
282
+ if (stream_object_signature(r, repl) < 0) {
283
error(_("hash mismatch %s"), oid_to_hex(oid));
284
return NULL;
285
}
@@ -290,7 +290,7 @@ struct object *parse_object(struct repository *r, const struct object_id *oid)
290
buffer = repo_read_object_file(r, oid, &type, &size);
291
if (buffer) {
292
if (check_object_signature(r, repl, buffer, size,
293
- type_name(type), NULL) < 0) {
293
+ type_name(type)) < 0) {
294
free(buffer);
295
error(_("hash mismatch %s"), oid_to_hex(repl));
296
return NULL;
pack-check.c
+6
-3
@@ -127,7 +127,7 @@ static int verify_packfile(struct repository *r,
127
128
if (type == OBJ_BLOB && big_file_threshold <= size) {
129
/*
130
- * Let check_object_signature() check it with
130
+ * Let stream_object_signature() check it with
131
* the streaming interface; no point slurping
132
* the data in-core only to discard.
133
*/
@@ -142,8 +142,11 @@ static int verify_packfile(struct repository *r,
142
err = error("cannot unpack %s from %s at offset %"PRIuMAX"",
143
oid_to_hex(&oid), p->pack_name,
144
(uintmax_t)entries[i].offset);
145
- else if (check_object_signature(r, &oid, data, size,
146
- type_name(type), NULL) < 0)
145
+ else if (data && check_object_signature(r, &oid, data, size,
146
+ type_name(type)) < 0)
147
+ err = error("packed %s from %s is corrupt",
148
+ oid_to_hex(&oid), p->pack_name);
149
+ else if (!data && stream_object_signature(r, &oid) < 0)
150
err = error("packed %s from %s is corrupt",
151
oid_to_hex(&oid), p->pack_name);
152
else if (fn) {