opensemantic.characteristics.quantitative

Library with Python models derived from the page package world.opensemantic.characteristics.quantitative

Quantities and units generated from QUDT. pint mapping was done with support from ucumvert

Usage

from opensemantic.characteristics.quantitative import (
    Length, LengthUnit, Width, Area, AreaUnit, QuantityValue,
)

# Create typed quantity values
length = Length(value=1.0, unit=LengthUnit.meter)
width = Width(value=200, unit=LengthUnit.milli_meter)

# Arithmetic — units are tracked automatically via pint
area = length * width
print(area.value, area.unit.name)  # 200000.0 milli_meter_squared
assert isinstance(area, Area)

# Unit conversion
area_cm2 = area.to_unit(AreaUnit.centi_meter_squared)
print(area_cm2.value)  # 2000.0
assert area == area_cm2  # physical equality across units

# Addition with automatic unit conversion
l1 = Length(value=1.0, unit=LengthUnit.milli_meter)
l2 = Length(value=1.0, unit=LengthUnit.meter)
print((l1 + l2).value)  # 1.001 (in meters)

# pint interop: round-trip to/from pint.Quantity
q_pint = l1.to_pint()                    # -> 1.0 millimeter
restored = QuantityValue.from_pint(q_pint)  # -> Length(1.0, milli_meter)
assert l1 == restored

# JSON serialization (via oold's unified API)
print(l1.to_json())
# {'type': ['Category:OSWee9c7e5c...'], 'value': 1.0, 'unit': 'Item:OSWf101d25e...'}

Round-tripping derived units

from_pint recovers a typed QuantityValue even when the exact unit symbol is not in the registry - e.g. derived units (watt, joule, ohm) or composed results such as volt * ampere. When the symbol is unknown it falls back to the quantity's dimensionality, mapping it to a representative SI-coherent unit via DIMENSION_TO_UNIT (generated into _dimensions.py by scripts/post_process.py):

from opensemantic.characteristics.quantitative import (
    Voltage, VoltageUnit, ElectricCurrent, ElectricCurrentUnit, Power,
)

power = Voltage(value=3.7, unit=VoltageUnit.volt) * ElectricCurrent(
    value=2.0, unit=ElectricCurrentUnit.ampere
)
assert isinstance(power, Power)  # volt * ampere -> watt

Compact export

exclude_defaults=True omits values left at their default (e.g. the default unit) for a minimal, portable representation:

length = Length(value=1.0, unit=LengthUnit.meter)
print(length.to_json(exclude_defaults=True))  # {'value': 1.0}

See examples/ for more, including tabular data with pandas and a tensile test analysis.

TODO: Unit Name Conflicts in _collection.py

18 unit member names appear in multiple *Unit enum classes with different OSW IDs. The first occurrence goes into _collection.py; conflicting entries remain hardcoded in _model.py. These likely stem from upstream QUDT modeling where the same unit name maps to different physical quantities.

Name	Classes (conflicting OSW IDs)
day	TimeUnit vs HydraulicPermeabilityUnit
kilo_newton	ForceUnit vs TorquePerLengthUnit
kilo_newton_per_meter	EnergyPerAreaUnit vs ForcePerLengthUnit
liter_per_meter_squared_per_second	VolumetricFluxUnit vs VentilationRatePerFloorAreaUnit
meter	LengthUnit vs AreaPerLengthUnit vs VolumePerUnitAreaUnit
meter_pascal	StressIntensityFactorUnit vs UnknownUnit
milli_newton_per_meter	EnergyPerAreaUnit vs ForcePerLengthUnit
minute	TimeUnit vs PlaneAngleUnit
newton	ForceUnit vs TorquePerLengthUnit
newton_per_meter	EnergyPerAreaUnit vs ForcePerLengthUnit
pH_value	InductanceUnit vs BasicityUnit
per_centi_meter	InverseLengthUnit vs UnknownUnit
per_day	FrequencyUnit vs UnknownUnit vs MassSpecificBiogeochemicalRateUnit
per_hour	FrequencyUnit vs MassSpecificBiogeochemicalRateUnit
per_meter	InverseLengthUnit vs UnknownUnit
per_meter_1	InverseLengthUnit vs UnknownUnit
per_minute	FrequencyUnit vs RotationalFrequencyUnit
year	AreaUnit vs TimeUnit

Status

pint round-trip (full inventory): 0 unit_registry misses

tests/conversion_test.py::test_full_inventory_test round-trips every QuantityValue x unit through to_pint() / from_pint() and now asserts zero unit_registry misses (v2: 2089 ok / 282 errors; v1: 3181 ok / 442 errors). from_pint resolves every unit whose symbol pint can parse - derived units fall back to the dimensionality map in _dimensions.py. The remaining errors are all UndefinedUnitError (units pint itself does not define, e.g. astronomical_unit, electron_volt, metric_ton, sidereal time units); the critical warnings are units shared by several quantities of the same dimensionality (value + unit still round-trip, but the class may resolve to a sibling).

Note

This project has been set up using PyScaffold 4.6. For details and usage information on PyScaffold see https://pyscaffold.org/.