builtin/cat-file: support "blob:limit=" objects filter
Implement support for the "blob:limit=" filter in git-cat-file(1), which causes us to omit all blobs that are bigger than a certain size. Signed-off-by: Patrick Steinhardt <ps@pks.im> Signed-off-by: Junio C Hamano <gitster@pobox.com>
Patrick Steinhardt committed
Apr 2, 2025 at 13:13 UTC
dbe1b32d59699092d549150e2db7af07e3cbfaf3
3 files changed
+34
-4
Documentation/git-cat-file.adoc
+5
@@ -91,6 +91,11 @@ OPTIONS
91
printed at all. The '<filter-spec>' may be one of the following:
92
+
93
The form '--filter=blob:none' omits all blobs.
94
++
95
+The form '--filter=blob:limit=<n>[kmg]' omits blobs of size at least n
96
+bytes or units. n may be zero. The suffixes k, m, and g can be used to name
97
+units in KiB, MiB, or GiB. For example, 'blob:limit=1k' is the same as
98
+'blob:limit=1024'.
99
100
--path=<path>::
101
For use with `--textconv` or `--filters`, to allow specifying an object
builtin/cat-file.c
+14
-1
@@ -483,8 +483,11 @@ static void batch_object_write(const char *obj_name,
483
int ret;
484
485
if (use_mailmap ||
486
- opt->objects_filter.choice == LOFC_BLOB_NONE)
486
+ opt->objects_filter.choice == LOFC_BLOB_NONE ||
487
+ opt->objects_filter.choice == LOFC_BLOB_LIMIT)
488
data->info.typep = &data->type;
489
+ if (opt->objects_filter.choice == LOFC_BLOB_LIMIT)
490
+ data->info.sizep = &data->size;
491
492
if (pack)
493
ret = packed_object_info(the_repository, pack, offset,
@@ -509,6 +512,15 @@ static void batch_object_write(const char *obj_name,
512
return;
513
}
514
break;
515
+ case LOFC_BLOB_LIMIT:
516
+ if (data->type == OBJ_BLOB &&
517
+ data->size >= opt->objects_filter.blob_limit_value) {
518
+ if (!opt->all_objects)
519
+ report_object_status(opt, obj_name,
520
+ &data->oid, "excluded");
521
+ return;
522
+ }
523
+ break;
524
default:
525
BUG("unsupported objects filter");
526
}
@@ -1049,6 +1061,7 @@ int cmd_cat_file(int argc,
1061
case LOFC_DISABLED:
1062
break;
1063
case LOFC_BLOB_NONE:
1064
+ case LOFC_BLOB_LIMIT:
1065
if (!batch.enabled)
1066
usage(_("objects filter only supported in batch mode"));
1067
break;
t/t1006-cat-file.sh
+15
-3
@@ -1356,11 +1356,12 @@ test_expect_success PERL '--batch-command info is unbuffered by default' '
1356
test_expect_success 'setup for objects filter' '
1357
git init repo &&
1358
(
1359
- # Seed the repository with three different sets of objects:
1359
+ # Seed the repository with four different sets of objects:
1360
#
1361
# - The first set is fully packed and has a bitmap.
1362
# - The second set is packed, but has no bitmap.
1363
# - The third set is loose.
1364
+ # - The fourth set is loose and contains big objects.
1365
#
1366
# This ensures that we cover all these types as expected.
1367
cd repo &&
@@ -1368,7 +1369,14 @@ test_expect_success 'setup for objects filter' '
1369
git repack -Adb &&
1370
test_commit second &&
1371
git repack -d &&
1371
- test_commit third
1372
+ test_commit third &&
1373
+
1374
+ for n in 1000 10000
1375
+ do
1376
+ printf "%"$n"s" X >large.$n || return 1
1377
+ done &&
1378
+ git add large.* &&
1379
+ git commit -m fourth
1380
)
1381
'
1382
@@ -1380,7 +1388,7 @@ test_expect_success 'objects filter with unknown option' '
1388
test_cmp expect err
1389
'
1390
1383
-for option in blob:limit=1 object:type=tag sparse:oid=1234 tree:1 sparse:path=x
1391
+for option in object:type=tag sparse:oid=1234 tree:1 sparse:path=x
1392
do
1393
test_expect_success "objects filter with unsupported option $option" '
1394
case "$option" in
@@ -1435,5 +1443,9 @@ test_objects_filter () {
1443
}
1444
1445
test_objects_filter "blob:none"
1446
+test_objects_filter "blob:limit=1"
1447
+test_objects_filter "blob:limit=500"
1448
+test_objects_filter "blob:limit=1000"
1449
+test_objects_filter "blob:limit=1k"
1450
1451
test_done