Dan Balasescu

### Type Game behaviour ### Bug description In osu!stable, Geki/Katu are used to denote when a strong hit occurred, depending on which of a GREAT/OK happened alongside. Geki: Strong hit...

Ranked convert map with low SV not converting correctly in taiko and mania

6

### Discussed in https://github.com/ppy/osu/discussions/17120 Originally posted by **zachmanthethird** March 6, 2022 Originally mentioned in #15747 and #15808 > https://osu.ppy.sh/b/1494828 At 02:32:949, there's a 0.01x velocity point with a slider that...

.NET analyser wish-list

8

I'm beginning to forget all the things we need analysers for, so will keep tracking them in this thread.

Bindable.BindTo() is not thread-safe

5

Calling `BindTo` on multiple threads can create the `Bindings` list multiple times: https://github.com/ppy/osu-framework/blob/ec4ba87c49f9a7b6756f62f7e1f664259123999a/osu.Framework/Bindables/Bindable.cs#L223-L227 ~~At one point this used to be a `Lazy`, but I think there was a reason for...

Remove local vertex storage from VertexBuffer

22

**This is an RFC. I will rebase and clean up this PR so please don't attempt to go through by-commit** @peppy @bdach @frenzibyte This PR is in a "final" state...

Space inserted before last character during Korean IME input

4

### Discussed in https://github.com/ppy/osu-framework/discussions/5045 Originally posted by **pnutbtr1** February 20, 2022 When entering Korean, spacing is placed before last character. I'm using Arch Linux with fcitx and X11. https://user-images.githubusercontent.com/94981525/154814822-0dc5dac4-b653-47ef-9c55-0131f01a3cb3.mp4 [logs.tar.gz](https://github.com/ppy/osu/files/8102818/logs.tar.gz)

`Bindable.Parse()` is currently culture-invariant, but can be used to parse user input. We should probably provide an overload of that function supporting a culture parameter. For example, see discussion https://github.com/ppy/osu/discussions/18271....

As of https://github.com/dotnet/runtime/pull/59514, the runtime implementation has been hoisted into the C# core library and no longer incurs allocation/performance overhead. It seems like this change is present in at least...

TextFlowContainer should wrap on individual characters as a last resort

5

In order to not overflow the parenting container, `TextFlowContainer` should be able to resort to wrapping on an individual character basis. This will need additional consideration for e.g. links (https://github.com/ppy/osu/issues/10085).

Lack of devirtualisation on Bindable<T>.Value

5

We are losing a considerable amount of performance due to `Bindable.Value` being `virtual`. Making `BindableInt` override + seal works however I'd rather push towards improving the bindable structure further instead...