Skip to main content

osu! difficulty and pp calculation for all modes

Project description

rosu-pp-py

Difficulty and performance calculation for all osu! modes.

This is a python binding to the Rust library rosu-pp which was bootstrapped through PyO3. Since all the heavy lifting is done by Rust, rosu-pp-py comes with a very fast performance. Check out rosu-pp's README for more info.

How to use rosu-pp-py

The library exposes four classes: Calculator, ScoreParams, CalculateResult, and Strains.

  1. The first step is to create a new Calculator instance by providing the constructor the path to a .osu beatmap file like so
calculator = Calculator('/path/to/file.osu')

Optionally, you can also provide the kwargs ar, cs, hp, or od to adjust the map's attributes or alternatively, after creating the calculator, you can call set_ar(v), set_cs(v), set_hp(v), or set_od(v).

calculator = Calculator('/path/to/file.osu', ar = 10.0)
calculator.set_od(9.2)
  1. Next, you need to create ScoreParams. It has the following fields:
mode: Optional[int],
    specify for scores on convert maps, default to the map's native mode
    available values are 0 for standard, 1 for taiko, 2 for catch, and 3 for mania
mods: Optional[int],
    bit value for mods, defaults to 0 (NM) see https://github.com/ppy/osu-api/wiki#mods
acc: Optional[float],
    if neither acc nor hitresults are specified, acc defaults to 100.0
n300: Optional[int],
    defaults to value based on acc
n100: Optional[int],
    defaults to value based on acc
n50: Optional[int],
    defaults to value based on acc
nMisses: Optional[int],
    defaults to 0
nKatu: Optional[int],
    only relevant for osu!ctb
combo: Optional[int],
    defaults to full combo
score: Optional[int],
    only relevant for osu!mania
passedObjects: Optional[int],
    only consider this many hit objects; useful for failed scores; defaults to all objects
clockRate: Optional[float]
    defaults to the mod's clock rate.

Note that all fields are optional. If nothing is specified, the parameters are equivalent to the parameters of the best possible NM score. ScoreParams can be created either by calling the constructor without arguments and then set the fields manually like so

params = ScoreParams()
params.acc = 98.76

or they can be created by passing kwargs to the constructor directly like so

params = ScoreParams(acc = 98.76)
  1. The last step is to provide the ScoreParams to the Calculator through the function calculate. This function takes one argument which must be either a single ScoreParams or an Iterable[ScoreParams], i.e. anything that python can iterate over like a list, set, ...

Example

from rosu_pp_py import Calculator, ScoreParams

calculator = Calculator('./maps/1980365.osu')

params1 = ScoreParams(
    mods = 8 + 16,  # HDHR
    acc = 97.89,
    nMisses = 13,
    combo = 1388,
)
params2 = ScoreParams(mods = 24)

# provide params for a single score, returns a list with one element
[result] = calculator.calculate(params1)

# provide multiple params
results = calculator.calculate([params1, params2])

assert result == results[0]

print(f'PP: {results[0].pp}/{results[1].pp} | Stars: {results[1].stars}')

Return object structure

The Calculator::calculate function will provide you a list of CalculateResult, one for each score you specified parameters for. CalculateResult contains the difficulty and performance attributes. Most of its attributes are optional based on the map's mode. In the following, O/T/C/M will denote for which mode the given attribute will be present:

mode: int
    Gamemode of the map, 0=O, 1=T, 2=C, 3=M. (O/T/C/M)
stars: float
    Star rating of the map. (O/T/C/M)
pp: float
    Performance points of the score. (O/T/C/M)
ppAcc: Optional[float]
    Accuracy based portion of the performance points. (O/T/M)
ppAim: Optional[float]
    Aim based portion of the performance points. (O)
ppFlashlight: Optional[float]
    Flashlight based portion of the performance points. (O)
ppSpeed: Optional[float]
    Speed based portion of the performance points. (O)
ppStrain: Optional[float]
    Strain based portion of the performance points. (T/M)
nFruits: Optional[int]
    The amount of fruits in the map. (C)
nDroplets: Optional[int]
    The amount of droplets in the map. (C)
nTinyDroplets: Optional[int]
    The amount of tiny droplets in the map. (C)
aimStrain: Optional[float]
    Aim based portion of the star rating. (O)
speedStrain: Optional[float]
    Speed based portion of the star rating. (O)
flashlightRating: Optional[float]
    Flashlight based portion of the star rating. (O)
sliderFactor: Optional[float]
    Nerf factor for sliders. (O)
ar: float
    Approach rate of the map. (O/T/C/M)
cs: float
    Circle size of the map. (O/T/C/M)
hp: float
    Health drain rate of the map. (O/T/C/M)
od: float
    Overall difficulty of the map. (O/T/C/M)
bpm: float
    Beats per minute of the map. (O/T/C/M)
clockRate: float
    Clock rate used in calculation i.e. 1.5 for DT, 0.75 for HT, 1.0 for NM or one that was specified (O/T/C/M)
timePreempt: Optional[float]
    The time in milliseconds in which the circles is visible before being clicked. (O)
greatHitWindow: Optional[float]
    The time in milliseconds in which a 300 ("great") is achievable. (O/T/M)
nCircles: Optional[int]
    The amount of circles in the map. (O/T/M)
nSliders: Optional[int]
    The amount of sliders in the map. (O/T/M)
nSpinners: Optional[int]
    The amount of spinners in the map. (O/T/C)
maxCombo: Optional[int]
    The max combo of the map. (O/T/C)

Calculating strains

If you want to plot the difficulty of a map over time, you can calculate the strain values. The return type of Calculator::strains is an instance of the Strains class. Its attributes depend on the map's game mode again and look as follows:

sectionLength: float
    The time in milliseconds between two strain values. (O/T/C/M)
aim: List[float]
    Strain values for the aim skill (O)
aimNoSliders: List[float]
    Strain values for the aim skill without sliders (O)
speed: List[float]
    Strain values for the speed skill (O)
flashlight: List[float]
    Strain values for the flashlight skill (O)
color: List[float]
    Strain values for the color skill (T)
rhythm: List[float]
    Strain values for the rhythm skill (T)
staminaLeft: List[float]
    Strain values for the left-stamina skill (T)
staminaRight: List[float]
    Strain values for the right-stamina skill (T)
strains: List[float]
    Strain values for the strain skill (M)
movement: List[float]
    Strain values for the movement skill (C)

Here's a small example

from rosu_pp_py import Calculator

calculator = Calculator('./maps/1980365.osu')
strains = calculator.strains(8 + 16) # HDHR
for i,strain in enumerate(strains.aim):
    currTime = i * strains.sectionLength
    print(f'Aim strain at {currTime}ms: {strain}')

Installing rosu-pp-py

Installing rosu-pp-py requires a supported version of Python and Rust.

Once Python and Rust and ready to go, you can install the project with pip:

$ pip install rosu-pp-py

or

$ pip install git+https://github.com/MaxOhn/rosu-pp-py

Learn More

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rosu_pp_py-0.8.0.tar.gz (14.8 kB view hashes)

Uploaded Source

Built Distributions

rosu_pp_py-0.8.0-pp37-pypy37_pp73-manylinux_2_5_x86_64.manylinux1_x86_64.whl (431.7 kB view hashes)

Uploaded PyPy manylinux: glibc 2.5+ x86-64

rosu_pp_py-0.8.0-pp37-pypy37_pp73-manylinux_2_5_i686.manylinux1_i686.whl (441.7 kB view hashes)

Uploaded PyPy manylinux: glibc 2.5+ i686

rosu_pp_py-0.8.0-cp310-none-win_amd64.whl (343.3 kB view hashes)

Uploaded CPython 3.10 Windows x86-64

rosu_pp_py-0.8.0-cp310-none-win32.whl (325.5 kB view hashes)

Uploaded CPython 3.10 Windows x86

rosu_pp_py-0.8.0-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.whl (429.1 kB view hashes)

Uploaded CPython 3.10 manylinux: glibc 2.5+ x86-64

rosu_pp_py-0.8.0-cp310-cp310-manylinux_2_5_i686.manylinux1_i686.whl (439.3 kB view hashes)

Uploaded CPython 3.10 manylinux: glibc 2.5+ i686

rosu_pp_py-0.8.0-cp39-none-win_amd64.whl (343.3 kB view hashes)

Uploaded CPython 3.9 Windows x86-64

rosu_pp_py-0.8.0-cp39-none-win32.whl (325.5 kB view hashes)

Uploaded CPython 3.9 Windows x86

rosu_pp_py-0.8.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.whl (429.1 kB view hashes)

Uploaded CPython 3.9 manylinux: glibc 2.5+ x86-64

rosu_pp_py-0.8.0-cp39-cp39-manylinux_2_5_i686.manylinux1_i686.whl (439.3 kB view hashes)

Uploaded CPython 3.9 manylinux: glibc 2.5+ i686

rosu_pp_py-0.8.0-cp39-cp39-macosx_10_9_x86_64.macosx_11_0_arm64.macosx_10_9_universal2.whl (775.9 kB view hashes)

Uploaded CPython 3.9 macOS 10.9+ universal2 (ARM64, x86-64) macOS 10.9+ x86-64 macOS 11.0+ ARM64

rosu_pp_py-0.8.0-cp39-cp39-macosx_10_7_x86_64.whl (404.1 kB view hashes)

Uploaded CPython 3.9 macOS 10.7+ x86-64

rosu_pp_py-0.8.0-cp38-none-win_amd64.whl (343.3 kB view hashes)

Uploaded CPython 3.8 Windows x86-64

rosu_pp_py-0.8.0-cp38-none-win32.whl (325.7 kB view hashes)

Uploaded CPython 3.8 Windows x86

rosu_pp_py-0.8.0-cp38-cp38-manylinux_2_5_x86_64.manylinux1_x86_64.whl (429.2 kB view hashes)

Uploaded CPython 3.8 manylinux: glibc 2.5+ x86-64

rosu_pp_py-0.8.0-cp38-cp38-manylinux_2_5_i686.manylinux1_i686.whl (439.4 kB view hashes)

Uploaded CPython 3.8 manylinux: glibc 2.5+ i686

rosu_pp_py-0.8.0-cp37-none-win_amd64.whl (343.3 kB view hashes)

Uploaded CPython 3.7 Windows x86-64

rosu_pp_py-0.8.0-cp37-none-win32.whl (325.7 kB view hashes)

Uploaded CPython 3.7 Windows x86

rosu_pp_py-0.8.0-cp37-cp37m-manylinux_2_5_x86_64.manylinux1_x86_64.whl (429.2 kB view hashes)

Uploaded CPython 3.7m manylinux: glibc 2.5+ x86-64

rosu_pp_py-0.8.0-cp37-cp37m-manylinux_2_5_i686.manylinux1_i686.whl (439.4 kB view hashes)

Uploaded CPython 3.7m manylinux: glibc 2.5+ i686

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page