fix multiplication of particles and quantity

[baggepinnen]

closes #152

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.33%. Comparing base (f4a9bd6) to head (26a6d10).
Report is 2 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #159      +/-   ##
==========================================
- Coverage   89.59%   89.33%   -0.27%     
==========================================
  Files          20       20              
  Lines        1153     1153              
==========================================
- Hits         1033     1030       -3     
- Misses        120      123       +3     

Flag	Coverage Δ
unittests	89.33% <100.00%> (-0.27%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[aplavin]

Seems to work for me, thanks!
Shouldn't introduce new ambiguities either...

[aplavin]

I wonder why these promote_op calls are needed at all?
Seems like

Base.$f(y::Quantity{S,D,U}, p::$PT{T,N}) where {S, D, U, T <: Number, N} = $PT($(op).(y, p.particles))

would work exactly the same.

[baggepinnen]

no, inference fails for this implementation, compare the two inference results below, where the first is with the implementation in this PR and the second is the one you propose

[aplavin]

I see, indeed – it cannot infer the 2nd argument, number of particles.
Still, it's generally recommended to avoid explicitly calling inference (promote_op) whenever possible, and the approach used elsewhere in this package is cleaner, eg

MonteCarloMeasurements.jl/src/register_primitive.jl

Lines 44 to 46 in 814ca57

	function ($m.$f)(p::$PT{T,N},a::Real...) where {T,N}
	res = ($m.$f).(p.particles, MonteCarloMeasurements.maybe_particles.(a)...) # maybe_particles introduced to handle >2 arg operators
	return $PT{eltype(res),N}(res)

:

function Base.$f(y::Quantity, p::$PT{T,N}) where {T, N}
    res = $(op).(y, p.particles)
    $PT{eltype(res), N}(res)
end