STUDIS Strongly Typed Units & Dimensions In SI

Copyright 2018 Morteza Jalalvand Licensed under the NDPL please see Licence for details.

Scientifically valid equations must be dimensionally homogeneous. It means that you can't compare quantities with different dimensions or add or subtract them. The argument of sine and many other mathematical functions must be a dimensionless quantity. Moreover, quantities of the same dimension but differing units should be converted to the same unit before comparing, adding or subtracting them. Breaking these rules in a program results in logical errors that can easily go undetected. STUDIS enforces the concept of dimensional homogeneity as syntax rules so that you get a compile error for violating it. It also internally converts all units to SI units so that quantities with differing units can be easily compared, added or subtracted.

Table of contents

• Usage
• Performance
• How many dimensions are there
• List of units
• List of constants
• Acknowledgement
• Dedication
• Licence

Usage

What you see in this section is basically the content of example.cpp.

You should begin by

#include"studis.hpp"usingnamespacestudis::literals;usingnamespacestudis::constants;

Then you can define and use quantities easily

auto l1 = 1.5_m, l2 = 2_cm;
auto t = 3_s;
auto l3 = l1 + l2; // fine
std::cout << l3 << std::endl; // prints '1.52 m' (yes the unit is printed as well)
std::cout << l1 + l2 << std::endl; // same
std::cout << (l1 < l2) << std::endl; // works// std::cout << l1 + t << std::endl; // error// std::cout << (l1 < t) << std::endl; // errorauto speed = l1 / t;
std::cout << speed << std::endl; // prints '0.5 m/s'

All math functions that make sense for quantities with dimension are overloaded

std::cout << abs (-1_A) << std::endl; // prints '1 A'
std::cout << atan2 (7_m, 1_km) << std::endl; // prints some number// std::cout << atan2 (1_m, 1_s) << std::endl;// error
std::cout << hypot (3_m, 4_m) << std::endl; // prints '5 m'// std::cout << hypot (1_m, 1_s) << std::endl;// error

pow is the only function that has different signature than its std counterpart, this can't be avoided since the dimension of the output depends on the power

auto energy = 0.5 * 1_kg * pow<2> (speed);
std::cout << energy << std::endl; // prints '0.125 J (m2.kg/s2)'

sqrt, cbrt are overloaded for quantities whose result does not have a non-integer dimensional exponent

// pi, standard_gravity and many other constants are defined in the constants namespaceauto pendulum_frequency = sqrt (standard_gravity / 1_m) / (2*pi);
std::cout << pendulum_frequency << std::endl; // prints '0.498403 Hz (1/s)'
std::cout << cbrt (1_litre) << std::endl; // prints '0.1 m'

Fractional power dimensions are not supported

// std::cout << sqrt (1_s) << std::endl; // error// std::cout << cbrt (1_m2) << std::endl; // error

Dimensionless quantities can be used with any math function since they implicitly convert to a floating-point

auto pos = 1_cm * cos (2*pi*1_s*pendulum_frequency);
std::cout << pos << std::endl; // prints '-0.0099995 m'// std::cout << cos (1_s) << std::endl; // error

There are so many units and prefixes in STUDIS

auto resistance = 1.7_kOhm; // we don't have greek letters, so that's kiloohmauto inductance = 1_uH; // same, this is microhenryauto capacitance = 1_pF;
if (resistance > 2*sqrt (inductance/capacitance))
std::cout << "overdamped" << std::endl;
elseif (resistance == 2*sqrt (inductance/capacitance))
std::cout << "cricitally damped" << std::endl;
else std::cout << "underdamped" << std::endl;

You can use STUDIS simply as a unit convertor (to SI units)

std::cout << 10_ly << std::endl; // prints '9.46073e+16 m'
std::cout << 30_knot << std::endl; // prints '15.4333 m/s'
std::cout << 1_MeV << std::endl; // prints '1.60218e-13 J (m2.kg/s2)'
std::cout << 2000_kcal << std::endl; // prints '8.368e+09 J (m2.kg/s2)'
std::cout << 120_mmHg << std::endl; // prints '15998.7 Pa (kg/m.s2)'

And so many constants

auto radiative_power = Stefan_Boltzmann_constant * pow<4>(300_K) * 1_m2;
std::cout << radiative_power << std::endl; // prints '459.3 W (m2.kg/s3)'
std::cout << electron_mass << std::endl; // prints '9.10938e-31 kg'

Value of a (non-const) variable can change but its dimension can't

auto mass = 1_kg;
mass = 300_g; // fine// mass = 1_m3; // error
std::cin >> mass; // you can also read its value
std::cout << mass << std::endl;

If you don't want to specify an initial value (not recommended), you have to specify the dimension of the quantity

studis::Density d;
std::cin >> d;
std::cout << d << std::endl;

Many common dimensions are there, but in the case you can't find it there, you can specify the power for all 7 base dimensions of the SI yourself

studis::Quantity<studis::Dimension<1,0,-3,0,0,0,0>> jerk;
std::cin >> jerk;
std::cout << jerk << std::endl;

Performance

STUDIS should not incur any noticeable overhead at runtime. Information about dimension of quantities are encoded in the type system so they are not stored and only the value itself consumes memory. All dimension checks are of course performed during compilation and incur no cost at runtime.

How many dimensions are there?

Really a lot. Much more than any reasonable use case scenario. The dimensional exponents of quantities can always range from -127 to 127 (it could actually be more), so at least about 2⁵⁶. In other words a quantity Q with dimension

dim Q = L^α M^β T^γ I^δ Θ^ε N^ζ J^η

is guaranteed to be in STUDIS as long as all of α, β, γ, δ, ε, ζ, and η are integers in interval -127 to 127.

Common dimensions have type-aliases for easy access

type-alias	dimension
Dimmensionless	1
Length	L
Mass	M
Time, Duration	T
ElectricCurrent	I
Temperature	Θ
AmountOfSubstance	N
LuminousIntensity	J
LuminousFlux	J
Wavenumber	L-1
Area	L2
Volume	L3
CurrentDensity	L-2 I
Density	L-3 M
Concentration	L-3 N
Velocity, Speed	L T-1
Acceleration	L T-2
Momentum	L M T-1
Action	L2 M T-1
Frequency	T-1
Radioactivity	T-1
Force	L M T-2
Pressure, Stress	L-1 M T-2
DynamicViscosity	L-1 M T-1
KinematicViscosity	L2 T-2
Torque	L2 M T-2
Energy, Work, Heat	L2 M T-2
Power, RadiantFlux	L2 M T-3
HeatCapacity	L2 M T-2 Θ-1
Entropy	L2 M T-2 Θ-1
ElectricCharge	T I
ElectricPotential, ElectromotiveForce, Voltage	L2 M T-3 I-1
Capacitance	L-2 M-1 T4 I2
Resistance, Impedance	L2 M T-3 I-2
Conductance, Admittance	L-2 M-1 T3 I2
MagneticFlux	L2 M T-2 I-1
MagneticFluxDensity	M T-3 I-1
Inductance	L2 M T-2 I-2
Illuminance	L-2 J
CatalyticActivity	T-1 N

List of units

Quantity	Unit	Symbols
Length	metre	fm, pm, nm, um, mm, cm, m, km, micron
Length	angstrom	angstrom
Length	inch	in
Length	foot	ft
Length	yard	yd
Length	mile	mile
Length	nautical mile	nautical_mile
Length	astronomical unit	au
Length	light year	ly, kly, Mly, Gly
Length	parsec	pc, kpc, Mpc, Gpc
Mass	gram	fg, pg, ng, ug, mg, g, gr, kg
Mass	dalton	Da, kDa, MDa
Mass	pound	lb
Mass	ounce	oz
Mass	tonne	t
Time	second	fs, ps, ns, us, ms, s, sec
Time	svedberg	Svedberg
Time	minute	min
Time	hour	h, hour
Time	day	d, day
Time	Julian year	julian_year
ElectricCurrent	ampere	nA, uA, mA, A, kA
Temperature	kelvin	K
Temperature	degree Celsius	deg_C, degree_Celsius
Temperature	degree Fahrenheit	deg_F, degree_Fahrenheit
AmountOfSubstance	mole	nmol, umol, mmol, mol, kmol
LuminousIntensity	candela	cd
Area		mm2, cm2, m2, km2
Area		in2, ft2, yd2, mile2
Area	barn	barn
Area	hectare	ha, hectare
Volume		cm3, m3
Volume	litre	ul, uL, ml, mL, l, L, litre
Density	gram per cubic centimetre	gr_per_cm3, gr_per_ml, gr_per_mL
Density	kilogram per litre	kg_per_l, kg_per_L
Density	kilogram per cubic metre	kg_per_m3
Concentration	molar	pM, nM, uM, mM, M
Velocity	metre per second	m_per_s
Velocity	foot per second	ft_per_s, ft_per_sec
Velocity	kilometre per hour	km_per_hour
Velocity	mile per hour	mile_per_hour
Velocity	knot	knot
Acceleration	metre per square second	m_per_s2
Acceleration	foot per square second	ft_per_s2
Acceleration	gal	Gal
Momentum	metre kilogram per second	m_kg_per_s
Action	joule second	J_s
Frequency	hertz	Hz, kHz, MHz, GHz, THz
Frequency	Baud	Bd, kBd, MBd, GBd
Frequency	FLOPS	FLOPS, kFLOPS, MFLOPS, GFLOPS, TFLOPS
Frequency	revolutions per minute	rpm
Frequency	frames per second	fps
Radioactivity	becquerel	Bq
Force	newton	pN, nN, uN, mN, N, kN
Force	dyne	dyn, dyne
Force	pound force	lbf
Pressure	pascal	Pa, kPa, MPa, GPa
Pressure	torr	mTorr, Torr
Pressure	millimetre of mercury	mmHg, cmHg
Pressure	psi	psi
Pressure	bar	mbar, bar
Pressure	standard atmosphere	atm
DynamicViscosity	pascal second	Pa_s
DynamicViscosity	poise	cP, P
KinematicViscosity	square metre per second	m2_per_s
KinematicViscosity	stokes	cSt, St
Torque	newton metre	N_m
Energy	joule	J, kJ, MJ, GJ
Energy	electronvolt	eV, keV, MeV, GeV
Energy	erg	erg
Energy	watt hour	Wh, kWh
Energy	british thermal unit	BTU
Energy	calorie	cal, kcal
Power	watt	nW, uW, mW, W, kW, MW, GW
ElectricCharge	coulomb	pC, nC, uC, mC, C
ElectricCharge	ampere hour	mAh, Ah
ElectricPotential	volt	uV, mV, V, kV, MV
Capacitance	farad	pF, nF, uF, mF, F
Resistance	ohm	uOhm, mOhm, Ohm, kOhm, MOhm, GOhm
Conductance	siemens	S
MagneticFlux	weber	nWb, uWb, mWb, Wb
MagneticFlux	maxwell	Mx
MagneticFluxDensity	tesla	uT, mT, T
MagneticFluxDensity	gauss	mG, G
Inductance	henry	uH, mH, H
Illuminance	lux	lx
CatalyticActivity	katal	kat

List of constants

Fundamental constants defining the 7 base units of the 2018 SI system

Constants	Defined value	Unit
speed_of_light	c = 299792458	m/s
Planck_constant	ℎ = 6.62607015 * 10-34	J s
elementary_charge	e = 1.602176634 * 10-19	C
Boltzmann_constant	k = 1.380649 * 10-23	J/K
Avogadro_constant	NA = 6.02214076 * 1023	1/mol
hyperfine_transition_frequency_of_Cs_133	ΔνCs = 9192631770	Hz
luminous_efficacy	Kcd = 873	lm/W

Fundamental constants whose values are exactly calculable in terms of the defined fundamental constants

Constants	Value	Unit
reduced_Planck_constant	ℏ = ℎ / (2 π)	J s
magnetic_flux_quantum	𝛷0 = ℎ / (2 e)	Wb
Josephson_constant	KJ = 2 e / ℎ	1/Wb
conductance_quantum	G0 = 2 e2 / ℎ	S
inverse_of_conductance_quantum	1 / G0	Ω
von_Klitzing_constant	RK = ℎ / e2	Ω
Faraday_constant	F = eNA	C/mol
molar_gas_constant, universal_gas_constant, gas_constant	R = kNA	J/(mol K)
Stefan_Boltzmann_constant	σ = (π2 / 60) k4 / (ℏ3c2)	W/(m2 K4)
first_radiation_constant	c1 = 2 π ℎ c2	W m2
second_radiation_constant	c2 = ℎ c / k	m K
Wien_displacement_law_constant, Wien_constant	b = 2.897771955185172... * 10-3	K m

Fundamental constants whose values are determined empirically

These values are based on the 2018 and 2019 set of values of the constants and conversion factors of physics and chemistry recommended by the Committee on Data for Science and Technology (CODATA).

Constants	Value	Unit	Relative standard uncertainty
magnetic_constant, vacuum_permeability	μ0 = 1.25663706212 * 10-6	N/A2	1.5 * 10-10
electric_constant, vacuum_permittivity	ε0 = 8.8541878128 * 10-12	F/m	1.5 * 10-10
characteristic_impedance_of_vacuum	Z0 = 376.730313668	Ω	1.5 * 10-10
Newtonian_constant_of_gravitation, universal_gravitational_constant, gravitational_constant	G = 6.67430 * 10-11	N/(m2 kg2)	2.2 * 10-5
atomic_mass_constant, atomic_mass_unit, Dalton	mu = 9.66053906660 * 10-27	kg	3.0 * 10-10
electron_mass	me = 9.1093837015 * 10-31	kg	3.0 * 10-10
proton_mass	mp = 1.67262192369 * 10-27	kg	3.1 * 10-10
proton_electron_mass_ratio	mp / me = 1836.15267343		6.0 * 10-11
fine_structure_constant	α = e2 / (4 π ε0 ℏ c) = 0.0072973525693		1.5 * 10-10
inverse_fine_structure_constant	α-1 = 137.035999084		1.5 * 10-10
Rydberg_constant	R∞ = α2mec / (2 ℎ) = 10973731.568160	1/m	1.9 * 10-12
Bohr_magneton	μB = e ℏ / (2 me) = 9.2740100783 * 10-24	J/T	3.0 * 10-10
nuclear_magneton	μB = e ℏ / (2 mp) = 5.0507837461 * 10-27	J/T	3.1 * 10-10
Bohr_radius	a0 = ℏ / (αme c) = 5.29177210903 * 10-11	m	1.5 * 10-10

Constants holding the value of non-SI units accepted for use with the International System of Units

Constant	Value
minute	1 min = 60 s
hour	1 h = 60 min = 3600 s
day	1 d = 24 h = 86400 s
degree	1° = (π/180) rad
arcminute	1′ = (1/60)° = (π/10800) rad
arcsecond	1″ = (1/60)′ = (π/648000) rad
hectare	1 ha = 104 m2
litre	1 L = 1 l = 10-3 m3
tonne	1 t = 103 kg

Constants holding the value of non-SI units associated with the CGS and the CGS-Gaussian system of units

Constant	Value
erg	1 erg = 10-7 J
dyne	1 dyn = 10-5 N
poise	1 P = 1 dyn s cm-2 = 0.1 Pa s
stokes	1 St = 1 cm2/s = 10-4 m2/s
gauss	1 G = 1 Mx/cm2 = 10-4 T
maxwell	1 Mx = 1 G cm2 = 10-8 Wb

Constants holding the value of non-SI units defined by the International Astronomical Union (IAU)

Constant	Value
julian_year	365.25 day
astronomical_unit	149597870700 m
light_year	Product of Julian year and speed of light
parsec	(648000/π) astronomical units

Adopted values

Constant	Value	Unit	Remarks
standard_gravity	gn = 9.80665	m/s2
standard_atmosphere	atm = 101325	Pa
standard_state_pressure	ssp = 100000	Pa
mercury_density	ρHg = 13595.1	kg/m3	Density used in the definition of mmHg

Constants holding the value of UK and US custmary units

Constant	Value
inch	1 in = 2.54 cm
foot	1 ft = 12 in
yard	1 yd = 3 ft
mile	1 mile = 1760 yd
nautical_mile	1 nautical mile = 1852 m
knot	1 knot = 1 nautical mile per hour
pound	1 lb = 0.45359237 kg
ounce	1 oz = (1/16) lb
pound_force	1 lbf = 1 lb * gn
pound_force_per_squared_inch	1 psi = 1 lbf/in2
british_thermal_unit	1 BTU = 788169 ft lbf
thermochemical_calorie	1 cal = 4184 J

Constants holding the value of other non-SI units

Constant	Value
angstrom	1 Å = 10-10 m
svedberg	1 S = 10-13 s
torr	1 Torr = (1/760) atm
millimeter_of_mercury	1 mmHg = ρHg * gn * 1 mm
watt_hour	1 Wh = 1 W * 1 h
ampere_hour	1 Ah = 1 A * 1 h

Acknowledgement

This is inspired by the idea of a strongly typed template MKS unit system discussed in the book The C++ Programming Language by Bjarne Stroustrup.

Dedication

This library is dedicated to all my mentors particularly Seyed Mehdi Vaez Allaei and Mohammad A. Charsooghi to whom I am grateful for both their teachings and friendship.

Licence

This library is distributed under the terms of Non-Discriminatory Public Licence. You can read the exact licence terms in the 'LICENSE' file, but here is a summary:

• You can use and modify the software
• You can distribute the original or the modified version of the software under the same terms in a non-discriminatory manner if you also provide the source code

If you have to comply with laws that compels you to restrict access of certain groups of people (such as export control laws), you can only use and modify this software for your own purposes, but you can no longer distribute it.