builtin/cat-file: wire up an option to filter objects
In batch mode, git-cat-file(1) enumerates all objects and prints them by iterating through both loose and packed objects. This works without considering their reachability at all, and consequently most options to filter objects as they exist in e.g. git-rev-list(1) are not applicable. In some situations it may still be useful though to filter objects based on properties that are inherent to them. This includes the object size as well as its type. Such a filter already exists in git-rev-list(1) with the `--filter=` command line option. While this option supports a couple of filters that are not applicable to our usecase, some of them are quite a neat fit. Wire up the filter as an option for git-cat-file(1). This allows us to reuse the same syntax as in git-rev-list(1) so that we don't have to reinvent the wheel. For now, we die when any of the filter options has been passed by the user, but they will be wired up in subsequent commits. Further note that the filters that we are about to introduce don't significantly speed up the runtime of git-cat-file(1). While we can skip emitting a lot of objects in case they are uninteresting to us, the majority of time is spent reading the packfile, which is bottlenecked by I/O and not the processor. This will change though once we start to make use of bitmaps, which will allow us to skip reading the whole packfile. 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
eb83e4c64b5a3458569593c2ab0c29365f10a82f
3 files changed
+88
-4
Documentation/git-cat-file.adoc
+16
@@ -81,6 +81,15 @@ OPTIONS
81
end-of-line conversion, etc). In this case, `<object>` has to be of
82
the form `<tree-ish>:<path>`, or `:<path>`.
83
84
+--filter=<filter-spec>::
85
+--no-filter::
86
+ Omit objects from the list of printed objects. This can only be used in
87
+ combination with one of the batched modes. Excluded objects that have
88
+ been explicitly requested via any of the batch modes that read objects
89
+ via standard input (`--batch`, `--batch-check`) will be reported as
90
+ "filtered". Excluded objects in `--batch-all-objects` mode will not be
91
+ printed at all. No filters are supported yet.
92
+
93
--path=<path>::
94
For use with `--textconv` or `--filters`, to allow specifying an object
95
name and a path separately, e.g. when it is difficult to figure out
@@ -340,6 +349,13 @@ the repository, then `cat-file` will ignore any custom format and print:
349
<object> SP missing LF
350
------------
351
352
+If a name is specified on stdin that is filtered out via `--filter=`,
353
+then `cat-file` will ignore any custom format and print:
354
+
355
+------------
356
+<object> SP excluded LF
357
+------------
358
+
359
If a name is specified that might refer to more than one object (an ambiguous short sha), then `cat-file` will ignore any custom format and print:
360
361
------------
builtin/cat-file.c
+32
-4
@@ -15,6 +15,7 @@
15
#include "gettext.h"
16
#include "hex.h"
17
#include "ident.h"
18
+#include "list-objects-filter-options.h"
19
#include "parse-options.h"
20
#include "userdiff.h"
21
#include "streaming.h"
@@ -35,6 +36,7 @@ enum batch_mode {
36
};
37
38
struct batch_options {
39
+ struct list_objects_filter_options objects_filter;
40
int enabled;
41
int follow_symlinks;
42
enum batch_mode batch_mode;
@@ -495,6 +497,13 @@ static void batch_object_write(const char *obj_name,
497
return;
498
}
499
500
+ switch (opt->objects_filter.choice) {
501
+ case LOFC_DISABLED:
502
+ break;
503
+ default:
504
+ BUG("unsupported objects filter");
505
+ }
506
+
507
if (use_mailmap && (data->type == OBJ_COMMIT || data->type == OBJ_TAG)) {
508
size_t s = data->size;
509
char *buf = NULL;
@@ -820,7 +829,8 @@ static int batch_objects(struct batch_options *opt)
829
struct object_cb_data cb;
830
struct object_info empty = OBJECT_INFO_INIT;
831
823
- if (!memcmp(&data.info, &empty, sizeof(empty)))
832
+ if (!memcmp(&data.info, &empty, sizeof(empty)) &&
833
+ opt->objects_filter.choice == LOFC_DISABLED)
834
data.skip_object_info = 1;
835
836
if (repo_has_promisor_remote(the_repository))
@@ -944,10 +954,13 @@ int cmd_cat_file(int argc,
954
int opt_cw = 0;
955
int opt_epts = 0;
956
const char *exp_type = NULL, *obj_name = NULL;
947
- struct batch_options batch = {0};
957
+ struct batch_options batch = {
958
+ .objects_filter = LIST_OBJECTS_FILTER_INIT,
959
+ };
960
int unknown_type = 0;
961
int input_nul_terminated = 0;
962
int nul_terminated = 0;
963
+ int ret;
964
965
const char * const builtin_catfile_usage[] = {
966
N_("git cat-file <type> <object>"),
@@ -1008,6 +1021,7 @@ int cmd_cat_file(int argc,
1021
N_("run filters on object's content"), 'w'),
1022
OPT_STRING(0, "path", &force_path, N_("blob|tree"),
1023
N_("use a <path> for (--textconv | --filters); Not with 'batch'")),
1024
+ OPT_PARSE_LIST_OBJECTS_FILTER(&batch.objects_filter),
1025
OPT_END()
1026
};
1027
@@ -1022,6 +1036,14 @@ int cmd_cat_file(int argc,
1036
if (use_mailmap)
1037
read_mailmap(&mailmap);
1038
1039
+ switch (batch.objects_filter.choice) {
1040
+ case LOFC_DISABLED:
1041
+ break;
1042
+ default:
1043
+ usagef(_("objects filter not supported: '%s'"),
1044
+ list_object_filter_config_name(batch.objects_filter.choice));
1045
+ }
1046
+
1047
/* --batch-all-objects? */
1048
if (opt == 'b')
1049
batch.all_objects = 1;
@@ -1076,7 +1098,8 @@ int cmd_cat_file(int argc,
1098
usage_msg_opt(_("batch modes take no arguments"),
1099
builtin_catfile_usage, options);
1100
1079
- return batch_objects(&batch);
1101
+ ret = batch_objects(&batch);
1102
+ goto out;
1103
}
1104
1105
if (opt) {
@@ -1108,5 +1131,10 @@ int cmd_cat_file(int argc,
1131
1132
if (unknown_type && opt != 't' && opt != 's')
1133
die("git cat-file --allow-unknown-type: use with -s or -t");
1111
- return cat_one_file(opt, exp_type, obj_name, unknown_type);
1134
+
1135
+ ret = cat_one_file(opt, exp_type, obj_name, unknown_type);
1136
+
1137
+out:
1138
+ list_objects_filter_release(&batch.objects_filter);
1139
+ return ret;
1140
}
t/t1006-cat-file.sh
+40
@@ -1353,4 +1353,44 @@ test_expect_success PERL '--batch-command info is unbuffered by default' '
1353
perl -e "$script" -- --batch-command $hello_oid "$expect" "info "
1354
'
1355
1356
+test_expect_success 'setup for objects filter' '
1357
+ git init repo
1358
+'
1359
+
1360
+test_expect_success 'objects filter with unknown option' '
1361
+ cat >expect <<-EOF &&
1362
+ fatal: invalid filter-spec ${SQ}unknown${SQ}
1363
+ EOF
1364
+ test_must_fail git -C repo cat-file --filter=unknown 2>err &&
1365
+ test_cmp expect err
1366
+'
1367
+
1368
+for option in blob:none blob:limit=1 object:type=tag sparse:oid=1234 tree:1 sparse:path=x
1369
+do
1370
+ test_expect_success "objects filter with unsupported option $option" '
1371
+ case "$option" in
1372
+ tree:1)
1373
+ echo "usage: objects filter not supported: ${SQ}tree${SQ}" >expect
1374
+ ;;
1375
+ sparse:path=x)
1376
+ echo "fatal: sparse:path filters support has been dropped" >expect
1377
+ ;;
1378
+ *)
1379
+ option_name=$(echo "$option" | cut -d= -f1) &&
1380
+ printf "usage: objects filter not supported: ${SQ}%s${SQ}\n" "$option_name" >expect
1381
+ ;;
1382
+ esac &&
1383
+ test_must_fail git -C repo cat-file --filter=$option 2>err &&
1384
+ test_cmp expect err
1385
+ '
1386
+done
1387
+
1388
+test_expect_success 'objects filter: disabled' '
1389
+ git -C repo cat-file --batch-check="%(objectname)" --batch-all-objects --no-filter >actual &&
1390
+ sort actual >actual.sorted &&
1391
+ git -C repo rev-list --objects --no-object-names --all >expect &&
1392
+ sort expect >expect.sorted &&
1393
+ test_cmp expect.sorted actual.sorted
1394
+'
1395
+
1396
test_done