Highlighted features

Dispatch on dimensions

Consider the following toy example, converting from voltage or power ratios to decibels:

julia> whatsit(x::Unitful.Voltage) = "voltage!"
whatsit (generic function with 1 method)

julia> whatsit(x::Unitful.Length) = "length!"
whatsit (generic function with 2 methods)

julia> whatsit(1u"mm")
"length!"

julia> whatsit(1u"kV")
"voltage!"

julia> whatsit(1u"A" * 2.5u"Ω")
"voltage!"

Dimensions in a type definition

It may be tempting to specify the dimensions of a quantity in a type definition, e.g.

struct Person
    height::Unitful.Length
    mass::Unitful.Mass
end

However, these are abstract types. If performance is important, it may be better just to pick a concrete Quantity type:

struct Person
    height::typeof(1.0u"m")
    mass::typeof(1.0u"kg")
end

You can still create a Person as Person(5u"ft"+10u"inch", 75u"kg"); the unit conversion happens automatically.

Making new units and dimensions

You can make new units using the @unit macro on the fly:

julia> @unit yd5 "yd5" FiveYards 5u"yd" false
yd5

Promotion is used to create arrays of a concrete type where possible, such that arrays of unitful quantities are stored efficiently in memory. However, if necessary, arrays can hold quantities with different dimensions, even mixed with unitless numbers. Doing so will suffer a performance penalty compared with the fast performance attainable with an array of concrete type (e.g. as resulting from [1.0u"m", 2.0u"cm", 3.0u"km"]). However, it could be useful in toy calculations for general relativity where some conventions yield matrices with mixed dimensions:

julia> Diagonal([-1.0u"c^2", 1.0, 1.0, 1.0])
4×4 Diagonal{Unitful.Quantity{Float64,D,U}}:
 -1.0 c^2   ⋅    ⋅    ⋅
       ⋅   1.0   ⋅    ⋅
       ⋅    ⋅   1.0   ⋅
       ⋅    ⋅    ⋅   1.0

Logarithmic units

julia> uconvert(u"mW*s", 20u"dBm/Hz")
100.0 s mW

Units with rational exponents

julia> 1.0u"V/sqrt(Hz)"
1.0 V Hz^-1/2

Exact conversions respected

julia> uconvert(u"ft",1u"inch")
1//12 ft