cat-file: sort and de-dup output of --batch-all-objects
The sorting we could probably live without, but printing duplicates is just a hassle for the user, who must then de-dup themselves (or risk a wrong answer if they are doing something like counting objects with a particular property). Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Jeff King committed
Jun 22, 2015 at 07:06 UTC
3115ee45c8c7c0b753663890b13ec0e14fe4c0d7
3 files changed
+18
-12
Documentation/git-cat-file.txt
+1
-2
@@ -74,8 +74,7 @@ OPTIONS
74
requested batch operation on all objects in the repository and
75
any alternate object stores (not just reachable objects).
76
Requires `--batch` or `--batch-check` be specified. Note that
77
- the order of the objects is unspecified, and there may be
78
- duplicate entries.
77
+ the objects are visited in order sorted by their hashes.
78
79
--buffer::
80
Normally batch output is flushed after each object is output, so
builtin/cat-file.c
+15
-7
@@ -9,6 +9,7 @@
9
#include "userdiff.h"
10
#include "streaming.h"
11
#include "tree-walk.h"
12
+#include "sha1-array.h"
13
14
struct batch_options {
15
int enabled;
@@ -324,19 +325,19 @@ struct object_cb_data {
325
struct expand_data *expand;
326
};
327
327
-static int batch_object_cb(const unsigned char *sha1,
328
- struct object_cb_data *data)
328
+static void batch_object_cb(const unsigned char sha1[20], void *vdata)
329
{
330
+ struct object_cb_data *data = vdata;
331
hashcpy(data->expand->sha1, sha1);
332
batch_object_write(NULL, data->opt, data->expand);
332
- return 0;
333
}
334
335
static int batch_loose_object(const unsigned char *sha1,
336
const char *path,
337
void *data)
338
{
339
- return batch_object_cb(sha1, data);
339
+ sha1_array_append(data, sha1);
340
+ return 0;
341
}
342
343
static int batch_packed_object(const unsigned char *sha1,
@@ -344,7 +345,8 @@ static int batch_packed_object(const unsigned char *sha1,
345
uint32_t pos,
346
void *data)
347
{
347
- return batch_object_cb(sha1, data);
348
+ sha1_array_append(data, sha1);
349
+ return 0;
350
}
351
352
static int batch_objects(struct batch_options *opt)
@@ -375,11 +377,17 @@ static int batch_objects(struct batch_options *opt)
377
data.info.typep = &data.type;
378
379
if (opt->all_objects) {
380
+ struct sha1_array sa = SHA1_ARRAY_INIT;
381
struct object_cb_data cb;
382
+
383
+ for_each_loose_object(batch_loose_object, &sa, 0);
384
+ for_each_packed_object(batch_packed_object, &sa, 0);
385
+
386
cb.opt = opt;
387
cb.expand = &data;
381
- for_each_loose_object(batch_loose_object, &cb, 0);
382
- for_each_packed_object(batch_packed_object, &cb, 0);
388
+ sha1_array_for_each_unique(&sa, batch_object_cb, &cb);
389
+
390
+ sha1_array_clear(&sa);
391
return 0;
392
}
393
t/t1006-cat-file.sh
+2
-3
@@ -548,7 +548,7 @@ test_expect_success 'git cat-file --batch --follow-symlink returns correct sha a
548
'
549
550
test_expect_success 'cat-file --batch-all-objects shows all objects' '
551
- # make new repos so we now the full set of objects; we will
551
+ # make new repos so we know the full set of objects; we will
552
# also make sure that there are some packed and some loose
553
# objects, some referenced and some not, and that there are
554
# some available only via alternates.
@@ -569,8 +569,7 @@ test_expect_success 'cat-file --batch-all-objects shows all objects' '
569
) >>expect.unsorted &&
570
sort <expect.unsorted >expect &&
571
git -C all-two cat-file --batch-all-objects \
572
- --batch-check="%(objectname)" >actual.unsorted &&
573
- sort <actual.unsorted >actual &&
572
+ --batch-check="%(objectname)" >actual &&
573
test_cmp expect actual
574
'
575