Dean Herbert
846291d203
Refactor new tests to not suck as much as the old importer tests
2022-07-26 16:33:00 +09:00
Dan Balasescu
412d42b907
Merge pull request #19393 from peppy/fix-statistics-panel
...
Avoid potential realm fetch after disposal in `StatisticsPanel`
2022-07-26 16:32:59 +09:00
Salman Ahmed
91ffa7007f
Improve existing test coverage to cover order changes from other sort modes
2022-07-26 10:24:51 +03:00
Salman Ahmed
a4f6f2b9eb
Make item ID comparer static
2022-07-26 09:55:37 +03:00
Dean Herbert
9939866f7d
Revert one more missed change
2022-07-26 15:54:10 +09:00
Dean Herbert
8370ca9765
Add ImportAsUpdate method to IModelImporter to avoid otehr changes
2022-07-26 15:49:04 +09:00
Salman Ahmed
693ac8750c
Remove remaining uses of "child" terminology in non-drawable components
2022-07-26 09:43:47 +03:00
Salman Ahmed
d7ef4170be
Maintain sort stability by using carousel item ID as fallback
2022-07-26 09:40:41 +03:00
Dean Herbert
4efe09e7c2
Merge pull request #19377 from frenzibyte/fix-timeline-tests
...
Fix intermittent timeline zoom test failures
2022-07-26 15:18:35 +09:00
Dan Balasescu
5345e431cc
Merge pull request #19389 from peppy/cursor-with-touch
...
Add ability to make cursor show even during touch input
2022-07-26 15:13:55 +09:00
Dean Herbert
99bdf41717
Avoid potential realm fetch after disposal in StatisticsPanel
...
As seen at https://github.com/ppy/osu/runs/7513799859?check_suite_focus=true .
2022-07-26 15:12:23 +09:00
Dean Herbert
ac553d22ea
Fix left over resource file which isn't actually imported
2022-07-26 15:08:18 +09:00
Dean Herbert
59d3cc52c4
Avoid leaving left-over files after test run completes
2022-07-26 15:08:18 +09:00
Dean Herbert
7d8a78ef01
Move tests to own class
2022-07-26 14:53:20 +09:00
Dan Balasescu
1b2158ff04
Remove unused method
2022-07-26 14:15:59 +09:00
Dean Herbert
ef10145d6f
Rename MenuCursorContainer and clean up code
2022-07-26 14:11:52 +09:00
Dean Herbert
ee0c67e114
Add ability to make cursor show even during touch input
...
I completely disagree with this from a UX perspective, but it's come up
so often that I figure we should just let users bone themselves.
2022-07-26 14:08:00 +09:00
Dan Balasescu
a7598c62ac
Merge pull request #19379 from frenzibyte/remove-submittable-score
...
Remove no longer necessary API score models
2022-07-26 14:07:40 +09:00
Andrew Hong
a8e315abf0
Refactor
2022-07-25 17:16:33 -04:00
Andrew Hong
46c4e78477
Add notification and another menuitem
2022-07-25 16:59:00 -04:00
Dean Herbert
d41ac36a69
Fix scenario where import is expected to be empty
2022-07-25 23:59:27 +09:00
Salman Ahmed
e5355f314d
Use longer hash string
2022-07-25 15:19:32 +03:00
Dean Herbert
9c411c2250
Fix test nullability assertions
2022-07-25 20:06:37 +09:00
Dean Herbert
8a0c8f5fd8
Fix some realm pieces not being cleaned up
2022-07-25 19:51:19 +09:00
Dean Herbert
92dd1bcddb
Add test coverage of actual update flow
2022-07-25 19:51:19 +09:00
Dean Herbert
2e14d8730c
Move implementation of updating a beatmap to BeatmapImporter
2022-07-25 19:51:19 +09:00
Dean Herbert
2363a3fb7b
Persist DateAdded over beatmap updates
2022-07-25 19:51:19 +09:00
Dean Herbert
e5ad07454c
Ensure previous version prior to update loses online info after marked pending delete
2022-07-25 19:51:19 +09:00
Dean Herbert
912218e123
Ensure scores are transferred after beatmap update if difficulty hash didn't change
2022-07-25 19:51:19 +09:00
Dean Herbert
b7f6413bce
Fix old version of beatmap potentially not being deleted during update flow
...
This can happen if the online IDs are not present in the `.osu` files.
Previously this was only working due to the early logic in the import
process (that relies on matching all online IDs perfectly).
2022-07-25 19:51:19 +09:00
Dean Herbert
6a3e8e31de
Centralise calls to reset online info of a BeatmapInfo
2022-07-25 19:51:19 +09:00
Dean Herbert
ec477a3ebf
Add basic coverage of current behaviour of beatmap reimport
2022-07-25 19:51:19 +09:00
Salman Ahmed
1b6ebcfd87
Remove SubmittableScore and replace with SoloScoreInfo extension method
2022-07-25 13:43:43 +03:00
Salman Ahmed
d04df19c7e
Remove APIScore and replace its final usage
2022-07-25 13:13:46 +03:00
Dan Balasescu
068063a43a
Merge pull request #19371 from peppy/fix-working-cache-invalidation
...
Fix calls to `GetWorkingBeatmap` invalidating cache too often
2022-07-25 18:29:46 +09:00
Dan Balasescu
1d48cee230
Merge pull request #19373 from peppy/fix-negative-star-rating-display
...
Fix star ratings which are still being calculated showing as "-1" at song select
2022-07-25 18:20:30 +09:00
Dan Balasescu
6937296192
Merge pull request #19374 from peppy/fps-counter-no-scheduler
...
Refactor `FPSCounter` to not use scheduled tasks
2022-07-25 18:19:59 +09:00
Salman Ahmed
48bcf57066
Mark SetupZoom and parameterless ZoomableScrollContainer ctor as protected
2022-07-25 12:07:50 +03:00
Salman Ahmed
bc2b629ee7
Let tests wait until track load before testing zoom
2022-07-25 12:07:50 +03:00
Salman Ahmed
07c6b44864
Fix Timeline attempting to setup zoom with unloaded track
2022-07-25 12:07:50 +03:00
Salman Ahmed
123930306b
Refactor ZoomableScrollContainer to allow setting up zoom range and initial zoom after load
2022-07-25 12:07:50 +03:00
Salman Ahmed
35462a04a0
Merge pull request #19367 from smoogipoo/log-diffcalc-failure
...
Log beatmap difficulty retrieval failures during score calculation
2022-07-25 11:49:55 +03:00
Salman Ahmed
93175eaf6e
Re-enable timeline zoom test and remove flaky attribute
2022-07-25 11:39:23 +03:00
Salman Ahmed
678acabf7e
Merge branch 'master' into log-diffcalc-failure
2022-07-25 11:16:31 +03:00
Andrew Hong
f1534da683
Formatting issues
2022-07-25 04:13:05 -04:00
Andrew Hong
fcf767e28b
Add contextmenu to beatmap external link
2022-07-25 04:07:33 -04:00
Dean Herbert
3c12b357e6
Merge pull request #19372 from frenzibyte/more-unprocessed-pp-handling
...
Update more places to support displaying unprocessed PP placeholder
2022-07-25 16:45:54 +09:00
Dean Herbert
2dd99ef1fd
Refactor FPSCounter to not use scheduled tasks
...
While on the surface this looks harmless (ignoring allocations),
`Scheduler` doesn't clear cancelled tasks until they reach their
execution time. This can cause an increase in time spent processing the
scheduler itself.
I don't think a per-frame updating component should use scheduled tasks
in this way in the first place, so I've just rewritten the logic to
avoid that overhead altogether.
2022-07-25 16:26:48 +09:00
Salman Ahmed
4d90e6bbac
Flip method to read better
2022-07-25 10:03:06 +03:00
Dean Herbert
6bf2645b1a
Fix StarRatingDisplay not handling negative numbers as "pending"
2022-07-25 15:44:54 +09:00
