Dean Herbert
bcdb73e1e8
Clean up files last
2021-11-25 14:28:27 +09:00
Dean Herbert
315e126497
Add automatic clean-up call for RealmFileStore
2021-11-25 14:17:44 +09:00
Dean Herbert
052ed8f930
Add online status to RealmBeatmapSet
2021-11-24 18:48:02 +09:00
Dan Balasescu
c4b5a07f4b
Merge branch 'master' into realm-move-data-migration-to-context-factory
2021-11-24 14:40:26 +09:00
Dan Balasescu
5a0a95f9d4
Trim whitespace
2021-11-24 13:48:59 +09:00
Dean Herbert
6aed41eacc
Merge branch 'master' into realm-move-data-migration-to-context-factory
2021-11-24 12:38:44 +09:00
Dean Herbert
63b09b356f
Rewrite commments to read better
2021-11-24 12:37:09 +09:00
Dean Herbert
6cd1bfd94f
Remove outdated comment
2021-11-24 12:35:05 +09:00
Bartłomiej Dach
0d409fa33e
Merge branch 'master' into realm-ruleset-keybinding-short-name
2021-11-23 20:18:58 +01:00
Dean Herbert
6b134359c9
Handle the case where the ruleset isn't found in the EF database any more
2021-11-23 19:15:52 +09:00
Dean Herbert
4149d458f0
Move realm data migrations inside the RealmContextFactory
2021-11-23 18:14:15 +09:00
Dean Herbert
d9917ee6c7
Merge branch 'realm-ruleset-setting-short-name' into realm-ruleset-keybinding-short-name
2021-11-23 18:13:26 +09:00
Dean Herbert
40cd998f99
Fix incorrect disposal
2021-11-23 18:13:05 +09:00
Dean Herbert
782191c410
Merge branch 'realm-ruleset-setting-short-name' into realm-ruleset-keybinding-short-name
2021-11-23 17:48:48 +09:00
Dean Herbert
b1b6723826
Add xmldoc and verbatim string markers
2021-11-23 17:47:43 +09:00
Dean Herbert
505fede44d
Pass the full EF context rather than a legacy RulesetStore
2021-11-23 17:40:20 +09:00
Dean Herbert
d94b27a8a2
Switch realm ruleset key bindings to use ruleset's ShortName as key
2021-11-22 18:52:30 +09:00
Dean Herbert
329bae50b0
Switch realm ruleset configuration to use ruleset's ShortName as key
2021-11-22 18:52:00 +09:00
Dean Herbert
ca26b6c540
Provide RealmContextFactory with the EF RulesetStore for migration purposes
2021-11-22 18:51:44 +09:00
Dean Herbert
d2062ff97f
Reformat realm migrations list for legibility
2021-11-22 18:47:18 +09:00
Dean Herbert
361cb78880
Fix realm applying migrations from one version too early
2021-11-22 18:46:46 +09:00
Dean Herbert
dfbc1f3394
Fix "conflicting" variable name
2021-11-09 17:42:03 +09:00
Dean Herbert
999d625e76
Fix realm migration potentially failing for users that haven't run osu! in a long time
...
As reported at https://github.com/ppy/osu/discussions/15530 .
2021-11-09 14:51:10 +09:00
Salman Ahmed
771bcbf868
Rename one more clashing variable
2021-11-05 12:24:07 +03:00
Dean Herbert
846c80f12c
Rename remaining clashing variables to appease CI somewhere
2021-11-05 17:02:24 +09:00
Dean Herbert
7772a97cf7
Add helper function to get realm class names rather than hard-coding each time
2021-11-05 17:01:00 +09:00
Dean Herbert
7acc4a4708
Refactor realm migrations to be a touch more legible
2021-11-04 18:57:54 +09:00
Dean Herbert
777b2cbcc4
Add realm migration for new author storage
2021-11-04 18:57:54 +09:00
Dean Herbert
3838fe5c6a
Fix typo in variable name
2021-10-29 11:39:15 +09:00
Dean Herbert
8d21f0b04b
Add migration to reset conflicting scroll speed key bindings
2021-10-29 11:14:22 +09:00
Dean Herbert
95837990f3
Apply some second-pass inspections that appeared after previous changes
2021-10-27 13:09:30 +09:00
Dean Herbert
6944151486
Apply batch fixing of built-in types using var
2021-10-27 13:04:41 +09:00
Dan Balasescu
08f3bc3f44
Merge pull request #15120 from peppy/realm-importer
...
Implement realm beatmap importer
2021-10-20 16:44:16 +09:00
Dean Herbert
d3ab45084d
Fix realm migration potentially failing from older releases
2021-10-19 14:19:58 +09:00
Dean Herbert
6904938dc1
Merge branch 'realm-indexed-online-id' into realm-importer
2021-10-18 16:14:37 +09:00
Dean Herbert
88a575462c
Work around weird null inspection
2021-10-18 16:11:55 +09:00
Dean Herbert
2c5ba1d8e2
Change OnlineID to non-nullable to allow for indexing in Realm
2021-10-18 15:53:40 +09:00
Dean Herbert
d95a62fa56
Add models and stores for beatmap manager requirements
2021-10-15 16:58:41 +09:00
Dean Herbert
818fac6ac8
Make realm a non-drawable component to better order disposal
...
Until now, the `RealmContextFactory` would be disposed as part of the
drawable hierarchy. This is too early, as it may be being used by higher
level components (like `ConfigManager`s, see #15115 ) that perform final
operations after the drawables have been disposed.
Seems to make sense moving this out of the drawable hierarchy and in
line with how we were doing things with EF.
2021-10-15 13:58:16 +09:00
Dean Herbert
f43badabf4
Add back update thread verification in RealmContextFactory
2021-10-11 15:20:12 +09:00
Dean Herbert
74841cf1a9
Merge branch 'update-realm-context-factory' into realm-context-factory-safer-blocking
2021-10-01 03:54:31 +09:00
Dean Herbert
b51fd00ba3
Guard against disposal in all context retrievals
2021-10-01 03:46:53 +09:00
Dean Herbert
8557530cd5
Add back main context locking
2021-10-01 03:45:00 +09:00
Dean Herbert
cfd3bdf888
Ensure realm blocks until all threaded usages are completed
2021-10-01 01:34:35 +09:00
Dean Herbert
9fa901f6aa
Refine RealmContext implementation API
2021-09-30 23:56:38 +09:00
Dean Herbert
f510ef9153
Move previousContext assign within lock to make things look safer
...
Not an actual requirement, but no harm.
2021-07-05 10:49:56 +09:00
Bartłomiej Dach
3291f15dcc
Mention GetForWrite() as another alternative to Context accesses
2021-07-04 12:08:15 +02:00
Dean Herbert
3ec7dc3bb9
Update tests in line with thread safety check
2021-07-04 17:59:39 +09:00
Dean Herbert
567e9f33a9
Fix thread safety of realm Refresh operation
...
Due to the lack of locking, there was a chance the the update thread
`context` was retrieved just before the `flushContexts` call, followed
by `.Refresh()` being run while the blocking behaviour was invoked.
This can be seen in test failures such as
https://ci.appveyor.com/project/peppy/osu/builds/39859786/tests .
As an aside, I tried multiple different methods to avoid `lock()` on the
update thread but they felt flaky. The overhead of lock when there's no
contention is reportedly around 30-50ns, so likely not of concern. We
can address it at a later point if it becomes one.
2021-07-04 16:24:43 +09:00
Dean Herbert
2f1203085b
Also add logging of realm block/flush operations
2021-06-29 20:23:25 +09:00
