@matheusbsbs / matheus-bagliano / commits / 8e0a1ec076

builtin/maintenance: introduce "reflog-expire" task

By default, git-maintenance(1) uses the "gc" task to ensure that the repository is well-maintained. This can be changed, for example by either explicitly configuring which tasks should be enabled or by using the "incremental" maintenance strategy. If so, git-maintenance(1) does not know to expire reflog entries, which is a subtask that git-gc(1) knows to perform for the user. Consequently, the reflog will grow indefinitely unless the user manually trims it. Introduce a new "reflog-expire" task that plugs this gap: - When running the task directly, then we simply execute `git reflog expire --all`, which is the same as git-gc(1). - When running git-maintenance(1) with the `--auto` flag, then we only run the task in case the "HEAD" reflog has at least N reflog entries that would be discarded. By default, N is set to 100, but this can be configured via "maintenance.reflog-expire.auto". When a negative integer has been provided we always expire entries, zero causes us to never expire entries, and a positive value specifies how many entries need to exist before we consider pruning the entries. Note that the condition for the `--auto` flags is merely a heuristic and optimized for being fast. This is because `git maintenance run --auto` will be executed quite regularly, so scanning through all reflogs would likely be too expensive in many repositories. Signed-off-by: Patrick Steinhardt <ps@pks.im> Signed-off-by: Junio C Hamano <gitster@pobox.com>

Patrick Steinhardt committed Apr 8, 2025 at 08:22 UTC 8e0a1ec0762405e045d924eed68b872fd29844c9
4 files changed +81
Documentation/config/maintenance.adoc
+9
@@ -69,3 +69,12 @@ maintenance.incremental-repack.auto::
69 Otherwise, a positive value implies the command should run when the
70 number of pack-files not in the multi-pack-index is at least the value
71 of `maintenance.incremental-repack.auto`. The default value is 10.
72 +
73 +maintenance.reflog-expire.auto::
74 + This integer config option controls how often the `reflog-expire` task
75 + should be run as part of `git maintenance run --auto`. If zero, then
76 + the `reflog-expire` task will not run with the `--auto` option. A
77 + negative value will force the task to run every time. Otherwise, a
78 + positive value implies the command should run when the number of
79 + expired reflog entries in the "HEAD" reflog is at least the value of
80 + `maintenance.loose-objects.auto`. The default value is 100.
Documentation/git-maintenance.adoc
+4
@@ -158,6 +158,10 @@ pack-refs::
158 need to iterate across many references. See linkgit:git-pack-refs[1]
159 for more information.
160
161 +reflog-expire::
162 + The `reflog-expire` task deletes any entries in the reflog older than the
163 + expiry threshold. See linkgit:git-reflog[1] for more information.
164 +
165 OPTIONS
166 -------
167 --auto::
builtin/gc.c
+50
@@ -33,6 +33,7 @@
33 #include "pack.h"
34 #include "pack-objects.h"
35 #include "path.h"
36 +#include "reflog.h"
37 #include "blob.h"
38 #include "tree.h"
39 #include "promisor-remote.h"
@@ -285,6 +286,49 @@ static int maintenance_task_pack_refs(struct maintenance_run_opts *opts,
286 return run_command(&cmd);
287 }
288
289 +struct count_reflog_entries_data {
290 + struct expire_reflog_policy_cb policy;
291 + size_t count;
292 + size_t limit;
293 +};
294 +
295 +static int count_reflog_entries(struct object_id *old_oid, struct object_id *new_oid,
296 + const char *committer, timestamp_t timestamp,
297 + int tz, const char *msg, void *cb_data)
298 +{
299 + struct count_reflog_entries_data *data = cb_data;
300 + if (should_expire_reflog_ent(old_oid, new_oid, committer, timestamp, tz, msg, &data->policy))
301 + data->count++;
302 + return data->count >= data->limit;
303 +}
304 +
305 +static int reflog_expire_condition(struct gc_config *cfg UNUSED)
306 +{
307 + timestamp_t now = time(NULL);
308 + struct count_reflog_entries_data data = {
309 + .policy = {
310 + .opts = REFLOG_EXPIRE_OPTIONS_INIT(now),
311 + },
312 + };
313 + int limit = 100;
314 +
315 + git_config_get_int("maintenance.reflog-expire.auto", &limit);
316 + if (!limit)
317 + return 0;
318 + if (limit < 0)
319 + return 1;
320 + data.limit = limit;
321 +
322 + repo_config(the_repository, reflog_expire_config, &data.policy.opts);
323 +
324 + reflog_expire_options_set_refname(&data.policy.opts, "HEAD");
325 + refs_for_each_reflog_ent(get_main_ref_store(the_repository), "HEAD",
326 + count_reflog_entries, &data);
327 +
328 + reflog_expiry_cleanup(&data.policy);
329 + return data.count >= data.limit;
330 +}
331 +
332 static int maintenance_task_reflog_expire(struct maintenance_run_opts *opts UNUSED,
333 struct gc_config *cfg UNUSED)
334 {
@@ -1383,6 +1427,7 @@ enum maintenance_task_label {
1427 TASK_GC,
1428 TASK_COMMIT_GRAPH,
1429 TASK_PACK_REFS,
1430 + TASK_REFLOG_EXPIRE,
1431
1432 /* Leave as final value */
1433 TASK__COUNT
@@ -1419,6 +1464,11 @@ static struct maintenance_task tasks[] = {
1464 maintenance_task_pack_refs,
1465 pack_refs_condition,
1466 },
1467 + [TASK_REFLOG_EXPIRE] = {
1468 + "reflog-expire",
1469 + maintenance_task_reflog_expire,
1470 + reflog_expire_condition,
1471 + },
1472 };
1473
1474 static int compare_tasks_by_selection(const void *a_, const void *b_)
t/t7900-maintenance.sh
+18
@@ -447,6 +447,24 @@ test_expect_success 'pack-refs task' '
447 test_subcommand git pack-refs --all --prune <pack-refs.txt
448 '
449
450 +test_expect_success 'reflog-expire task' '
451 + GIT_TRACE2_EVENT="$(pwd)/reflog-expire.txt" \
452 + git maintenance run --task=reflog-expire &&
453 + test_subcommand git reflog expire --all <reflog-expire.txt
454 +'
455 +
456 +test_expect_success 'reflog-expire task --auto only packs when exceeding limits' '
457 + git reflog expire --all --expire=now &&
458 + test_commit reflog-one &&
459 + test_commit reflog-two &&
460 + GIT_TRACE2_EVENT="$(pwd)/reflog-expire-auto.txt" \
461 + git -c maintenance.reflog-expire.auto=3 maintenance run --auto --task=reflog-expire &&
462 + test_subcommand ! git reflog expire --all <reflog-expire-auto.txt &&
463 + GIT_TRACE2_EVENT="$(pwd)/reflog-expire-auto.txt" \
464 + git -c maintenance.reflog-expire.auto=2 maintenance run --auto --task=reflog-expire &&
465 + test_subcommand git reflog expire --all <reflog-expire-auto.txt
466 +'
467 +
468 test_expect_success '--auto and --schedule incompatible' '
469 test_must_fail git maintenance run --auto --schedule=daily 2>err &&
470 test_grep "at most one" err