Merge pull request #368 from FourierFlows/ncc/aliasing-docs

Fix figures in Docs/Aliasing & Docs/Timestepping + add \citet ref style

Project.toml

@@ -20,7 +20,7 @@ Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 
 [compat]
-CUDA = "1, 2.4.2, 3.0.0 - 3.6.4, 3.7.1, 4, 5"
+CUDA = "1, 2.4.2, 3.0.0 - 3.6.4, 3.7.1, 4, 5.4"
 DocStringExtensions = "0.8, 0.9"
 FFTW = "1"
 Interpolations = "0.12, 0.13, 0.14, 0.15"

docs/make.jl

@@ -10,7 +10,7 @@ using
 #####
 
 bib_filepath = joinpath(@__DIR__, "src/references.bib")
-bib = CitationBibliography(bib_filepath)
+bib = CitationBibliography(bib_filepath, style=:authoryear)
 
 const EXAMPLES_DIR = joinpath(@__DIR__, "..", "examples")
 const OUTPUT_DIR   = joinpath(@__DIR__, "src/literated")

docs/src/aliasing.md

@@ -46,9 +46,9 @@ lines!(ax, range(-Lx/2, Lx/2, length=200), f2;
        color = (:salmon, 0.3), label = "cos(6x)")
 
 scatter!(ax, x, f1.(x);
-         markersize = 12, color = (:steelblue, 0.9), label = "cos(4x)")
+         markersize = 24, marker = :star5, color = (:steelblue, 0.9), label = "cos(4x)")
 scatter!(ax, x, f2.(x);
-         markersize = 24, marker = :star5, color = (:salmon, 0.9), label = "cos(6x)")
+         markersize = 12, color = (:salmon, 0.9), label = "cos(6x)")
 
 axislegend(merge = true)
 
@@ -82,15 +82,21 @@ harmonics with wavenumbers beyond those that our grid is able to resolve and, th
 aliasing errors.
 
 The above-mentioned 1/2-rule for dealiasing for quadratic nonlinearities is, however, not the 
-most efficient. [Orszag-1971](@cite) pointed out that for quadratic nonlirearities, simply only
+most efficient. [Orszag-1971](@citet) pointed out that for quadratic nonlirearities, simply only
 discarding the highest-1/3 of wavenumber components is enough to save us from aliasing errors.
 To be fair, with Orszag's so-called 2/3-rule for dealiasing, still some aliasing errors occur, but
 those  errors only occur to the higher-1/3 wavenumber components that will be zero-ed out at the
 next time step, when we next dealias our solution anyway.
 
 When constructing a `grid` we can specify the `aliased_fraction` parameter. By default, this is 
-set to ``1/3``, appropriate for quadratic nonlinearities. Then `dealias!(fh, grid)` will zero-out 
-the highest-`aliased_fraction` wavenumber components of `fh`. 
+set to ``1/3``, appropriate for quadratic nonlinearities.
+
+```@example 1
+grid = OneDGrid(; nx, Lx)
+```
+
+Then we can use `dealias!(fh, grid)` to zero-out the highest-`aliased_fraction` wavenumber
+components of `fh`.
 
 ```@docs
 FourierFlows.dealias!

docs/src/timestepping.md

@@ -5,7 +5,7 @@ FourierFlows.jl includes several time-stepping algorithms.
 Most of the time-stepping algorithms are fully explicit schemes: [`ForwardEulerTimeStepper`](@ref),
 [`AB3TimeStepper`](@ref), [`RK4TimeStepper`](@ref), and [`LSRK54TimeStepper`](@ref).
 Also we have implemented an [`ETDRK4TimeStepper`](@ref) scheme with the improvements described
-by [Kassam-Trefethen-2005](@cite).
+by [Kassam-Trefethen-2005](@citet).
 
 The [`Problem`](@ref FourierFlows.Problem) constructor expects the chosen time stepper as
 as string that includes the corresponding name of the time stepper _without_ the ending `TimeStepper`.
@@ -41,9 +41,9 @@ close to machine precision. That is:
 \alpha = \frac{- \log\delta}{(k_{\textrm{max}} - k_{\textrm{cutoff}})^p} \ .
 ```
 
-The above-mentioned filter form originates from the book by [Canuto-etal-1987](@cite).
-In geophysical turbulence applications it was used by [LaCasce-1996](@cite) and later
-by [Arbic-Flierl-2004](@cite).
+The above-mentioned filter form originates from the book by [Canuto-etal-1987](@citet).
+In geophysical turbulence applications it was used by [LaCasce-1996](@citet) and later
+by [Arbic-Flierl-2004](@citet).
 
 !!! warning "Not too steep, not too shallow"
     Care should be taken if one decides to fiddle with the filter parameters. Changing
@@ -70,15 +70,15 @@ using FourierFlows: makefilter
 K = 0:0.001:1 # non-dimensional wavenumber k * dx / π
 
 fig = Figure()
-ax = Axis(fig[1, 1], xlabel = "|𝐤| dx / π", ylabel = "filter", aspect=2.5, xticks=0:0.2:1)
+ax = Axis(fig[1, 1], xlabel = "|𝐤| dx / π", ylabel = "filter", xticks=0:0.2:1)
 
 vlines!(ax, 2/3, color = (:gray, 0.4), linewidth = 6, label = "cutoff wavenumber |𝐤| dx / π = 2/3 (default)")
 
 lines!(ax, K, makefilter(K), linewidth = 4, label = "order 4 (default)")
 lines!(ax, K, makefilter(K, order = 1), linestyle = :dash, label = "order 1")
 lines!(ax, K, makefilter(K, order = 10), linestyle = :dot, label = "order 10")
 
-axislegend(position = :lb)
+fig[0, 1] = Legend(fig, ax)
 
 current_figure() # hide
 ```

src/timesteppers.jl

@@ -306,7 +306,7 @@ uⁿ⁺¹ = uⁿ
 ```
 
 where `Aᵢ`, `Bᵢ`, and `Cᵢ` are the ``A``, ``B``, and ``C`` coefficients from
-the LSRK table at the ``i``-th stage. For details, refer to [Carpenter-Kennedy-1994](@cite).
+the LSRK table at the ``i``-th stage. For details, refer to [Carpenter-Kennedy-1994](@citet).
 
 !!! info "Usage"
     The `LSRK54TimeStepper` is *slower* than the [`RK4TimeStepper`](@ref) but
@@ -429,7 +429,7 @@ That is,
 uⁿ⁺¹ = exp(L * dt) * uⁿ + RK4(N(uⁿ))
 ```
 
-For more info refer to [Kassam-Trefethen-2005](@cite).
+For more info refer to [Kassam-Trefethen-2005](@citet).
 """
 struct ETDRK4TimeStepper{T,TL} <: AbstractTimeStepper{T}
   # ETDRK4 coefficents
@@ -684,7 +684,7 @@ Calculate the coefficients associated with the (diagonal) linear coefficient
 `L` for an ETDRK4 timestepper with timestep `dt`.
 
 The calculation is done by integrating over a unit circle in the complex space.
-For more info refer to [Kassam-Trefethen-2005](@cite).
+For more info refer to [Kassam-Trefethen-2005](@citet).
 """
 function getetdcoeffs(dt, L; ncirc=32, rcirc=1)
   shape = Tuple(cat(ncirc, ones(Int, ndims(L)), dims=1))