1522 Commits

• Mark spinner rewind test as flaky
  Dan Balasescu · 2025-09-23 18:18:07 +09:00
  ff6c6083b4
• Replace old-skin with retro skin in tests
  clayton · 2025-09-05 21:21:52 +09:00
  3c1c537b1b
• Remove no longer valid test
  Salman Alshamrani · 2025-08-23 09:28:14 +03:00
  c0c3690908
• Merge pull request #34606 from cl8n/sliderpoint

  Add skin support for sliderpoint10 and sliderpoint30

  Bartłomiej Dach · 2025-08-19 15:06:20 +02:00
  47fecfb669
• Fix manual click test being broken
  Bartłomiej Dach · 2025-08-18 13:46:06 +02:00
  df210241fc
• Improve test coverage somewhat
  Bartłomiej Dach · 2025-08-18 13:20:38 +02:00
  d696ac99d4
• Rename everything to start with
  Bartłomiej Dach · 2025-08-18 12:46:53 +02:00
  777ab61143
• Merge branch 'master' into hitposition
  Bartłomiej Dach · 2025-08-18 12:20:51 +02:00
  fe612d465b
• Don't switch on array index in slidertick test
  clayton · 2025-08-16 09:32:31 -07:00
  e77fb987a9
• Don't use switch expression
  clayton · 2025-08-16 05:08:30 -07:00
  c4163e33e5
• Fix Slider case of test
  clayton · 2025-08-16 04:57:05 -07:00
  a96d00a55f
• Fix failing test
  Bartłomiej Dach · 2025-08-12 14:26:00 +02:00
  0cd4fbb416
• Add tick judgement gallery test scene
  clayton · 2025-08-10 07:16:18 -07:00
  cd7a304640
• Add failing test case
  Salman Alshamrani · 2025-08-05 14:20:20 +03:00
  5aa186c987
• Merge remote-tracking branch 'upstream/master' into hitposition
  cdwcgt · 2025-08-03 00:35:54 +08:00
  0fa0568f13
• Refactor beatmap attribute methods

  This change refactors `GetAdjustedDisplayDifficulty()` and
  `GetBeatmapAttributesToDisplay()` in two ways:
  
  - Both methods now accept `IBeatmapInfo` instead of
    `IBeatmapDifficultyInfo`. This is done in order to make mania key
    count display to work, wherein `IBeatmapDifficultyInfo` is not enough
    to calculate the final key count.
  
  - `GetAdjustedDisplayDifficulty()` now applies all
    `IApplicableToDifficulty` mods itself. I did this after noticing that
    every real consumer of this method had to do that themselves for very
    little reason.

  Bartłomiej Dach · 2025-07-30 14:06:34 +02:00
  7c4dd812b6
• Attempt to properly quantify the impact of mania Hard Rock / Easy mod application on overall difficulty

  In stable mania, Hard Rock and Easy mods do not work the same way as
  they do on all of the rulesets. The difference is that mania HR and EZ,
  rather than apply a multiplier to the map's original Overall Difficulty,
  apply multipliers to *the durations of hit windows themselves*.
  
  Prior to the last release, lazer was oblivious to this reality and just
  treated mania HR / EZ as it did every other ruleset. Last release, for
  the sake for gameplay parity across rulesets, the mods in question were
  adjusted to match stable, but in the process, it started looking like HR
  / EZ did not change OD anymore.
  
  The problem is that they do, but applying a multiplier to the map's OD
  and applying a multiplier to the hit window duration is not the same
  thing. The second thing is actually *much harsher* in magnitude, to the
  point where applying HR to any map is almost guaranteed to exceed "the
  effective OD" of 10, and applying EZ to any map is almost guaranteed to
  result in "negative effective OD".
  
  This change attempts to convey that reality by displaying "effective
  OD", similar to what's already done in other rulesets when rate-changing
  mods are active. Note that the values this will display *do not match*
  stable *and that is correct*, because stable song select *lies* about
  the actual impact on OD by just assuming it can treat all rulesets in
  the same way.
  
  ---
  
  Would close https://github.com/ppy/osu/issues/34150 I guess.
  
  And yes I would like *all of the above* to land on the changelog if
  possible if this is merged.
  
  For further convincing that this makes any semblance of sense please see
  the following: https://www.desmos.com/calculator/yigt7jycdv

  Bartłomiej Dach · 2025-07-15 09:40:55 +02:00
  131f828e6a
• Mark TestSpinPerMinuteOnRewind as flaky

  See: https://github.com/ppy/osu/actions/runs/16167239898/job/45631919718

  Dan Balasescu · 2025-07-09 20:30:29 +09:00
  4f56099694
• Merge pull request #33910 from bdach/dont-do-a-barrel-roll-actually

  Fix Flashlight having increased radius when Barrel Roll is active

  Dean Herbert · 2025-07-02 18:25:45 +09:00
  0e39e05d19
• Merge pull request #33882 from bdach/hitwindow-halves

  Apply flooring and half-millisecond-adjustments to hit windows

  Dan Balasescu · 2025-07-02 18:09:32 +09:00
  9047d27197
• Merge pull request #33908 from bdach/custom-sample-bank-buggery

  Fix beatmap skin sample lookups falling back to non-custom sample banks if the custom bank sample was not found

  Dean Herbert · 2025-06-30 14:32:21 +09:00
  95237ed531
• Adjust tests to pass (and add test coverage of fail case)
  Bartłomiej Dach · 2025-06-27 14:07:00 +02:00
  edafac2aaa
• Make Flashlight test case exercising playfield scaling not useless

  It was doing a "if I touch the game in this very specific manner
  everything works" which will light up in very nice green colours but is
  actually useless for preventing regressions.

  Bartłomiej Dach · 2025-06-27 13:48:42 +02:00
  19abd4fbca
• Fix flaky tests due to async disposal
  Dan Balasescu · 2025-06-27 19:43:26 +09:00
  c53a7aa2fc
• Simplify important frame management
  Dan Balasescu · 2025-06-25 21:50:04 +09:00
  5dc08d4351
• Add test hitting next circle during tail window
  Dan Balasescu · 2025-06-25 21:20:00 +09:00
  5d69af55f0
• Uncomment & adjust relevant replay test cases

  The replay stability tests needed adjustments because hit windows have
  been materially changed with the previous commit. What matters in the
  replay stability tests is covering the time instants near the hit window
  edges and ensuring that re-encode doesn't mutate the resulting
  judgements, not what the particular numbers used are.

  Bartłomiej Dach · 2025-06-25 12:22:19 +02:00
  89c48451cd
• Apply flooring and half-millisecond-adjustments to hit windows

  This is a "two-birds-with-one-stone" change, which addresses both
  https://github.com/ppy/osu/issues/28744 and
  https://github.com/ppy/osu/issues/11311 simultaneously.
  
  - The replay stability issue caused by time instants being rounded to
    nearest integer is fixed by this, because flooring and
    subtracting/adding 0.5 from the hit window threshold makes it
    impossible for a judgement to switch to anything else after replay
    rounding is applied - all hit windows are always a full integer plus
    0.5 milliseconds, which immunizes them to rounding-to-full-ms issues.
  
  - The direction of applying the 0.5 adjustment additionally fixes the
    disparity with stable - in osu! and taiko 0.5 is subtracted as
    hit window ranges in those rulesets are exclusive on stable, while in
    mania 0.5 is added, as the hit window ranges there are *inclusive* on
    stable.
  
  As should be obvious, this materially changes hit windows. To what degree
  this is a *significant* change is up for discussion; I would say "no"
  since hitting half a millisecond changes would require 2000fps input
  recording, and we're still timestamping inputs using the update thread's
  clock, that gives a 1ms resolution at best.
  
  In the worst case, in osu! and taiko, this can change a hit window range
  by 1.5ms (e.g. 300.9ms -> floored to 300ms -> 299.5ms after subtraction
  of the half). It's more than the best-case resolution of input
  timestamps, but not by much. Considering how cleanly this resolves the
  issues in question, I see it as an acceptable tradeoff.

  Bartłomiej Dach · 2025-06-25 11:44:13 +02:00
  0f078ee550
• Refactor hit windows class structure to reduce rigidity

  This change pulls back a significant degree of overspecialisation and
  rigidity in the class structure of `HitWindows` to make subsequent
  changes to hit windows, whose purpose is to improve replay playback
  accuracy, possible to do cleanly.
  
  Notably:
  
  - `HitWindows` is full abstract now. In a few use cases, and as a
    reference for ruleset implementors, `DefaultHitWindows` is provided as
    a separate class instead.
  
    This fixes the weirdness wherein `HitWindows` always declared 6 fields
    for result types but some of them would never be set to a non-zero
    value or read.
  
  - `HitWindow.GetRanges()` is deleted because it is overspecialised and
    prevents being able to adjust hitwindows by ±0.5ms cleanly which will
    be required later.
  
    The fallout of this is that the assertion that used `GetRanges()` in
    the `HitWindows` ctor must use something else now, and the closest
    thing to it was `GetAllAvailableWindows()`, which didn't return
    the miss window - so I made it return the miss window and fixed the
    one consumer that didn't want it (bar hit error meter) to skip it.
  
  - Diff also contains some clean-up around `DifficultyRange` to unify
    handling of it.

  Bartłomiej Dach · 2025-06-25 08:40:12 +02:00
  8b542e5442
• Adjust test timings to avoid flaky test failing

  The gameplay clock runs at 1000 ms intervals, and the previous duration
  meant that the "store" step could cause a missed spinning check in a bad
  case scenario.

  Dean Herbert · 2025-06-19 19:16:12 +09:00
  088659ae04
• Add failing test
  Bartłomiej Dach · 2025-06-06 09:08:26 +02:00
  f577cea715
• Mark some more recent flaky tests
  Dean Herbert · 2025-06-03 16:23:31 +09:00
  ef29eda3e0
• Use a fixed clock
  Bartłomiej Dach · 2025-05-23 08:29:45 +02:00
  44feb7814e
• Do not use unseeded RNG in replay analysis container test

  It's just bad form.

  Bartłomiej Dach · 2025-05-22 12:52:40 +02:00
  d7efce5378
• Add failing test case
  Bartłomiej Dach · 2025-05-20 14:17:15 +02:00
  b2de486aec
• Adjust failing replay recording tests

  cd4ab9307c made these tests fail because
  they were basically pressing and releasing a key in a single frame,
  which hopefully never actually happens in real scenarios. (It's a bit
  tricky to determine if it can.)

  Bartłomiej Dach · 2025-05-16 08:59:56 +02:00
  953a93309c
• Add baseline test coverage of replay recording in all rulesets

  Nothing fancy, just basics to ensure the subsequent commit doesn't break
  anything.

  Bartłomiej Dach · 2025-05-14 11:25:14 +02:00
  168c3cf018
• Add test coverage for osu! hit windows with various mods active
  Bartłomiej Dach · 2025-04-21 12:54:26 +02:00
  cff16ed029
• Fix code quality once more
  Bartłomiej Dach · 2025-04-16 08:35:35 +02:00
  8273583fd0
• Add test cases covering correct legacy replay playback with respect to hitwindow treatment

  This continues on https://github.com/ppy/osu/pull/32770 via adding test
  cases which cover treatment of hit windows in stable in osu!, taiko, and
  mania. The test cases are exportable to beatmap `.osu` files and replay
  `.osr` files for stable crosscheck by setting `ExportLocation` on the
  test scene classes to a non-null path.
  
  For the most part, osu! and taiko ground truth matches previous findings
  - hit windows in those rulesets are floored to the nearest integer.
  The real "star" of this diff is mania, because:
  
  - The hit windows in mania depend on:
    - overall difficulty (as expected)
    - whether Score V2 is active
    - if Score V2 is not active, the hit windows also depend on whether
      the map was converted from another ruleset or not
  
  - Regardless of all aforementioned factors, mania hitwindows are *not
    symmetrical*. Due to what *appears* to be a straight-up bug, it is
    *not possible to achieve a MEH / 50 hit result when hitting late*.
    There is specific code that coerces late hits beyond 100 hit window
    range to full misses:
  
       https://github.com/peppy/osu-stable-reference/blob/996648fba06baf4e7d2e0b248959399444017895/osu!/GameplayElements/HitObjectManagerMania.cs#L737-L751
  
  Note that despite the fact that I'm PRing these test cases, none of this
  is a promise that all of stable behaviours will be returning unchanged
  when I PR something to actually do something about this and the other
  issue of replay instability. This is just coverage, to be used for
  awareness of what's still broken. The extent of how much stable is going
  to be humored here going forward will be subject to negotiation.

  Bartłomiej Dach · 2025-04-15 13:22:06 +02:00
  82b2a92894
• Add extra comments regarding notation
  Bartłomiej Dach · 2025-04-14 13:13:21 +02:00
  9cfba9008f
• Fix code quality
  Bartłomiej Dach · 2025-04-11 14:16:24 +02:00
  d4cae3232b
• Abstractify test scene

  I think in this case it's genuinely reasonable to use abstracts to
  reduce boilerplate.

  Bartłomiej Dach · 2025-04-11 12:09:53 +02:00
  30bb8e335c
• Add test exercising osu! replay playback stability after encode
  Bartłomiej Dach · 2025-04-11 12:09:51 +02:00
  ac36e228b8
• Avoid large sample depool overhead on drawable hitobjects with many nested objects

  Closes https://github.com/ppy/osu/issues/32588.

  Dean Herbert · 2025-03-26 18:45:12 +09:00
  78de898f7f
• Move BeatmapVersion from BeatmapInfo to IBeatmap

  Closes https://github.com/ppy/osu/issues/32420.
  
  The failure cause here is that in editor the beatmap version for the
  beatmap affected (or... any beatmap, really), is 0 (ZERO). That is
  probably a regression from https://github.com/ppy/osu/pull/32315, but
  like... can we universally agree that calling that change "a regression"
  in any capacity is dumb? Like what was that code *doing* playing dumb
  reference games and copying stuff into an arbitrary instance that could
  get or not get used later on? And now you have a 50/50 chance of
  accessing the *correct* model's field, depending on whether you go via
  `BeatmapInfo` or `Beatmap.BeatmapInfo`?
  
  Moving the field to `IBeatmap`, i.e. what is by now - by consensus,
  since https://github.com/ppy/osu/pull/28473 - supposed to be the "decoded
  and materialised" beatmap, fixes this issue.
  
  I probably should have done this as part of
  https://github.com/ppy/osu/pull/28473 but it slipped my mind. Probably
  for the better too because this change has rather large chances of
  breaking stuff so maybe better to examine it in isolation (via diffcalc
  runs or whatever).
  
  For added humour points, you'd say that the field on `BeatmapInfo` was
  not `[Ignore]`d, so this is a realm schema change, right? No. As far as
  I can tell, it's not. I opened realm studio and `BeatmapVersion` *is not
  a listed column` on `Beatmap` models.
  
  I'm also not gonna get into the fact that I think `EditorBeatmap` doing
  dumb games with juggling two `BeatmapInfo` references since
  https://github.com/ppy/osu/pull/15075 is bad, because I don't think I
  have the mental capacity to hotfix this by going down that train of
  thought.

  Bartłomiej Dach · 2025-03-17 15:04:48 +01:00
  91f3be5fea
• Fix placement blueprint tests
  Bartłomiej Dach · 2025-03-11 07:55:48 +01:00
  c962210b4f
• Merge branch 'master' into pp-dev
  Dean Herbert · 2025-02-27 00:08:43 +09:00
  87b6dddd11
• Merge pull request #31232 from Plextora/sudden-death-feat

  Add "Fail when missing on a slider end" option in Sudden Death mod

  Dean Herbert · 2025-02-18 14:25:46 +09:00
  f5bdbb91d6
• Add test coverage
  Dean Herbert · 2025-02-18 13:08:33 +09:00
  2d8e35be32