Merge branch 'trixi-framework:main' into main

trixi-framework · Jun 22, 2023 · 47a4f69 · 47a4f69
2 parents 41451f9 + bea4bfe
commit 47a4f69
Show file tree

Hide file tree

Showing 92 changed files with 1,852 additions and 776 deletions.
diff --git a/.github/workflows/SpellCheck.yml b/.github/workflows/SpellCheck.yml
@@ -10,4 +10,4 @@ jobs:
       - name: Checkout Actions Repository
         uses: actions/checkout@v3
       - name: Check spelling
-        uses: crate-ci/typos@v1.14.12
+        uses: crate-ci/typos@v1.15.1
diff --git a/NEWS.md b/NEWS.md
@@ -9,6 +9,7 @@ for human readability.
 #### Added
 
 - Experimental support for 3D parabolic diffusion terms has been added.
+- Capability to set truly discontinuous initial conditions in 1D.
 
 #### Changed
 

diff --git a/Project.toml b/Project.toml
@@ -1,7 +1,7 @@
 name = "Trixi"
 uuid = "a7f1ee26-1774-49b1-8366-f1abc58fbfcb"
 authors = ["Michael Schlottke-Lakemper <[email protected]>", "Gregor Gassner <[email protected]>", "Hendrik Ranocha <[email protected]>", "Andrew R. Winters <[email protected]>", "Jesse Chan <[email protected]>"]
-version = "0.5.29-pre"
+version = "0.5.30-pre"
 
 [deps]
 CodeTracking = "da1fd8a2-8d9e-5ec2-8556-3022fb5608a2"

diff --git a/docs/literate/src/files/adding_nonconservative_equation.jl b/docs/literate/src/files/adding_nonconservative_equation.jl
@@ -147,7 +147,7 @@ plot(sol)
 # above.
 
 error_1 = analysis_callback(sol).l2 |> first
-@test isapprox(error_1, 0.0002961027497) #src
+@test isapprox(error_1, 0.00029609575838969394) #src
 # Next, we increase the grid resolution by one refinement level and run the
 # simulation again.
 

diff --git a/docs/src/github-git.md b/docs/src/github-git.md
@@ -112,7 +112,7 @@ branch, and the corresponding pull request will be updated automatically.
 Please note that a review has nothing to do with the lack of experience of the
 person developing changes: We try to review all code before it gets added to
 `main`, even from the most experienced developers. This is good practice and
-helps to keep the error rate low while ensuring the the code is developed in a
+helps to keep the error rate low while ensuring that the code is developed in a
 consistent fashion. Furthermore, do not take criticism of your code personally -
 we just try to keep Trixi.jl as accessible and easy to use for everyone.
 
@@ -121,7 +121,7 @@ Once your branch is reviewed and declared ready for merging by the reviewer,
 make sure that all the latest changes have been pushed. Then, one of the
 developers will merge your PR. If you are one of the developers, you can also go
 to the pull request page on GitHub and and click on **Merge pull request**.
-Voilá, you are done! Your branch will have been merged to
+Voilà, you are done! Your branch will have been merged to
 `main` and the source branch will have been deleted in the GitHub repository
 (if you are not working in your own fork).
 

diff --git a/docs/src/styleguide.md b/docs/src/styleguide.md
@@ -53,9 +53,9 @@ of your PR), you need to install JuliaFormatter.jl first by running
 ```shell
 julia -e 'using Pkg; Pkg.add("JuliaFormatter")'
 ```
-You can then recursively format all Julia files in the Trixi.jl repo by executing
+You can then recursively format the core Julia files in the Trixi.jl repo by executing
 ```shell
-julia -e 'using JuliaFormatter; format(".")
+julia -e 'using JuliaFormatter; format(["benchmark", "ext", "src", "utils"])'
 ```
 from inside the Trixi.jl repository. For convenience, there is also a script you can
 directly run from your terminal shell, which will automatically install JuliaFormatter in a
@@ -65,3 +65,14 @@ utils/trixi-format.jl
 ```
 You can get more information about using the convenience script by running it with the
 `--help`/`-h` flag.
+
+### Checking formatting before committing
+It can be convenient to check the formatting of source code automatically before each commit. 
+We use git-hooks for it and provide a `pre-commit` script in the `utils` folder. The script uses
+[JuliaFormatter.jl](https://github.com/domluna/JuliaFormatter.jl) just like formatting script that 
+runs over the whole Trixi.jl directory. 
+You can copy the `pre-commit`-script into `.git/hooks/pre-commit` and it will check your formatting 
+before each commit. If errors are found the commit is aborted and you can add the corrections via
+```shell 
+git add -p
+```
diff --git a/docs/src/visualization.md b/docs/src/visualization.md
@@ -339,7 +339,7 @@ create a [`PlotData1D`](@ref) with the keyword argument `curve` set to your list
 Let's give an example of this with the basic advection equation from above by creating
 a plot along the circle marked in green:
 
-![2d-plot-along-cirlce](https://user-images.githubusercontent.com/72009492/130951042-e1849447-8e55-4798-9361-c8badb9f3a49.png)
+![2d-plot-along-circle](https://user-images.githubusercontent.com/72009492/130951042-e1849447-8e55-4798-9361-c8badb9f3a49.png)
 
 We can write a function like this, that outputs a list of points on a circle:
 ```julia

diff --git a/examples/dgmulti_3d/elixir_advection_tensor_wedge.jl b/examples/dgmulti_3d/elixir_advection_tensor_wedge.jl
@@ -0,0 +1,56 @@
+using OrdinaryDiffEq
+using Trixi
+using LinearAlgebra
+
+###############################################################################
+equations = LinearScalarAdvectionEquation3D(1.0, 1.0, 1.0)
+
+initial_condition = initial_condition_convergence_test
+
+# Define the polynomial degrees for the polynoms of the triangular base and the line
+# of the tensor-prism
+tensor_polydeg = (3, 4)
+
+dg = DGMulti(element_type = Wedge(),
+             approximation_type = Polynomial(),
+             surface_flux = flux_lax_friedrichs,
+             polydeg = tensor_polydeg)
+
+
+cells_per_dimension = (8, 8, 8)
+mesh = DGMultiMesh(dg, 
+                   cells_per_dimension,
+                   coordinates_min = (-1.0, -1.0, -1.0), 
+                   coordinates_max = (1.0, 1.0, 1.0),
+                   periodicity = true)
+
+
+semi = SemidiscretizationHyperbolic(mesh, equations, initial_condition, dg,
+                boundary_conditions=boundary_condition_periodic)
+
+###############################################################################
+# ODE solvers, callbacks etc.
+
+tspan = (0.0, 5.0)
+ode = semidiscretize(semi, tspan)
+
+summary_callback = SummaryCallback()
+
+analysis_interval = 100
+analysis_callback = AnalysisCallback(semi, interval=analysis_interval, uEltype=real(dg))
+
+alive_callback = AliveCallback(analysis_interval=analysis_interval)
+
+# The StepsizeCallback handles the re-calculation of the maximum Δt after each time step
+stepsize_callback = StepsizeCallback(cfl=1.0)
+
+callbacks = CallbackSet(summary_callback, analysis_callback, alive_callback, stepsize_callback)
+
+
+###############################################################################
+# run the simulation
+
+sol = solve(ode, CarpenterKennedy2N54(williamson_condition = false), dt = 1.0,
+            save_everystep=false, callback=callbacks);
+
+summary_callback() # print the timer summary
diff --git a/examples/p4est_2d_dgsem/elixir_euler_sedov.jl b/examples/p4est_2d_dgsem/elixir_euler_sedov.jl
@@ -11,7 +11,7 @@ equations = CompressibleEulerEquations2D(1.4)
     initial_condition_sedov_blast_wave(x, t, equations::CompressibleEulerEquations2D)
 
 The Sedov blast wave setup based on Flash
-- http://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114000000000000000
 """
 function initial_condition_sedov_blast_wave(x, t, equations::CompressibleEulerEquations2D)
   # Set up polar coordinates
@@ -20,7 +20,7 @@ function initial_condition_sedov_blast_wave(x, t, equations::CompressibleEulerEq
   y_norm = x[2] - inicenter[2]
   r = sqrt(x_norm^2 + y_norm^2)
 
-  # Setup based on http://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel/node184.html#SECTION010114000000000000000
+  # Setup based on https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114000000000000000
   r0 = 0.21875 # = 3.5 * smallest dx (for domain length=4 and max-ref=6)
   E = 1.0
   p0_inner = 3 * (equations.gamma - 1) * E / (3 * pi * r0^2)

diff --git a/examples/p4est_2d_dgsem/elixir_euler_supersonic_cylinder.jl b/examples/p4est_2d_dgsem/elixir_euler_supersonic_cylinder.jl
@@ -3,7 +3,7 @@
 # Boundary conditions are supersonic Mach 3 inflow at the left portion of the domain
 # and supersonic outflow at the right portion of the domain. The top and bottom of the
 # channel as well as the cylinder are treated as Euler slip wall boundaries.
-# This flow results in strong shock refletions / interactions as well as Kelvin-Helmholtz
+# This flow results in strong shock reflections / interactions as well as Kelvin-Helmholtz
 # instabilities at later times as two Mach stems form above and below the cylinder.
 #
 # For complete details on the problem setup see Section 5.7 of the paper:

diff --git a/examples/p4est_3d_dgsem/elixir_euler_sedov.jl b/examples/p4est_3d_dgsem/elixir_euler_sedov.jl
@@ -11,7 +11,7 @@ equations = CompressibleEulerEquations3D(1.4)
     initial_condition_medium_sedov_blast_wave(x, t, equations::CompressibleEulerEquations3D)
 
 The Sedov blast wave setup based on Flash
-- http://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114000000000000000
 with smaller strength of the initial discontinuity.
 """
 function initial_condition_medium_sedov_blast_wave(x, t, equations::CompressibleEulerEquations3D)
@@ -22,7 +22,7 @@ function initial_condition_medium_sedov_blast_wave(x, t, equations::Compressible
   z_norm = x[3] - inicenter[3]
   r = sqrt(x_norm^2 + y_norm^2 + z_norm^2)
 
-  # Setup based on http://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel/node184.html#SECTION010114000000000000000
+  # Setup based on https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114000000000000000
   r0 = 0.21875 # = 3.5 * smallest dx (for domain length=4 and max-ref=6)
   E = 1.0
   p0_inner = 3 * (equations.gamma - 1) * E / (4 * pi * r0^2)

diff --git a/examples/paper_self_gravitating_gas_dynamics/elixir_eulergravity_jeans_instability.jl b/examples/paper_self_gravitating_gas_dynamics/elixir_eulergravity_jeans_instability.jl
@@ -15,7 +15,7 @@ The classical Jeans instability taken from
 - Dominik Derigs, Andrew R. Winters, Gregor J. Gassner, Stefanie Walch (2016)
   A Novel High-Order, Entropy Stable, 3D AMR MHD Solver with Guaranteed Positive Pressure
   [arXiv: 1605.03572](https://arxiv.org/abs/1605.03572)
-- Flash manual https://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel.pdf
+- Flash manual https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node189.html#SECTION010131000000000000000
 in CGS (centimeter, gram, second) units.
 """
 function initial_condition_jeans_instability(x, t,
@@ -32,7 +32,7 @@ function initial_condition_jeans_instability(x, t,
   pres0  = 1.5e7 # dyn/cm^2
   delta0 = 1e-3
   # set wave vector values for perturbation (units 1/cm)
-  # see FLASH manual: https://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel.pdf
+  # see FLASH manual: https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node189.html#SECTION010131000000000000000
   kx = 2.0*pi/0.5 # 2π/λ_x, λ_x = 0.5
   ky = 0.0   # 2π/λ_y, λ_y = 1e10
   k_dot_x = kx*x[1] + ky*x[2]
@@ -49,7 +49,7 @@ function initial_condition_jeans_instability(x, t,
                                              equations::HyperbolicDiffusionEquations2D)
   # gravity equation: -Δϕ = -4πGρ
   # Constants taken from the FLASH manual
-  # https://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel.pdf
+  # https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node189.html#SECTION010131000000000000000
   rho0   = 1.5e7
   delta0 = 1e-3
 

diff --git a/examples/paper_self_gravitating_gas_dynamics/elixir_eulergravity_sedov_blast_wave.jl b/examples/paper_self_gravitating_gas_dynamics/elixir_eulergravity_sedov_blast_wave.jl
@@ -15,14 +15,14 @@ Adaptation of the Sedov blast wave with self-gravity taken from
   A purely hyperbolic discontinuous Galerkin approach for self-gravitating gas dynamics
   [arXiv: 2008.10593](https://arxiv.org/abs/2008.10593)
 based on
-- http://flash.uchicago.edu/site/flashcode/user_support/flash4_ug_4p62/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114100000000000000
 Should be used together with [`boundary_condition_sedov_self_gravity`](@ref).
 """
 function initial_condition_sedov_self_gravity(x, t, equations::CompressibleEulerEquations2D)
   # Set up polar coordinates
   r = sqrt(x[1]^2 + x[2]^2)
 
-  # Setup based on http://flash.uchicago.edu/site/flashcode/user_support/flash4_ug_4p62/node184.html#SECTION010114000000000000000
+  # Setup based on https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114100000000000000
   r0 = 0.125 # = 4.0 * smallest dx (for domain length=8 and max-ref=8)
   E = 1.0
   p_inner   = (equations.gamma - 1) * E / (pi * r0^2)
@@ -59,7 +59,7 @@ Adaptation of the Sedov blast wave with self-gravity taken from
   A purely hyperbolic discontinuous Galerkin approach for self-gravitating gas dynamics
   [arXiv: 2008.10593](https://arxiv.org/abs/2008.10593)
 based on
-- http://flash.uchicago.edu/site/flashcode/user_support/flash4_ug_4p62/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114100000000000000
 Should be used together with [`initial_condition_sedov_self_gravity`](@ref).
 """
 function boundary_condition_sedov_self_gravity(u_inner, orientation, direction, x, t,
@@ -122,7 +122,7 @@ Adaptation of the Sedov blast wave with self-gravity taken from
   A purely hyperbolic discontinuous Galerkin approach for self-gravitating gas dynamics
   [arXiv: 2008.10593](https://arxiv.org/abs/2008.10593)
 based on
-- http://flash.uchicago.edu/site/flashcode/user_support/flash4_ug_4p62/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114100000000000000
 Should be used together with [`boundary_condition_sedov_self_gravity`](@ref).
 """
 function initial_condition_sedov_self_gravity(x, t, equations::HyperbolicDiffusionEquations2D)
@@ -143,7 +143,7 @@ Adaptation of the Sedov blast wave with self-gravity taken from
   A purely hyperbolic discontinuous Galerkin approach for self-gravitating gas dynamics
   [arXiv: 2008.10593](https://arxiv.org/abs/2008.10593)
 based on
-- http://flash.uchicago.edu/site/flashcode/user_support/flash4_ug_4p62/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114100000000000000
 Should be used together with [`initial_condition_sedov_self_gravity`](@ref).
 """
 function boundary_condition_sedov_self_gravity(u_inner, orientation, direction, x, t,

diff --git a/examples/structured_1d_dgsem/elixir_advection_shockcapturing.jl b/examples/structured_1d_dgsem/elixir_advection_shockcapturing.jl
@@ -9,7 +9,9 @@ advection_velocity = 1.0
 """
     initial_condition_composite(x, t, equations::LinearScalarAdvectionEquation1D)
 
-Slight simplification of
+Wave form that is a combination of a Gaussian pulse, a square wave, a triangle wave,
+and half an ellipse with periodic boundary conditions.
+Slight simplification from
 - Jiang, Shu (1996)
   Efficient Implementation of Weighted ENO Schemes
   [DOI: 10.1006/jcph.1996.0130](https://doi.org/10.1006/jcph.1996.0130)
@@ -60,7 +62,7 @@ volume_integral = VolumeIntegralShockCapturingHG(indicator_sc;
 solver = DGSEM(basis, surface_flux, volume_integral)
 
 # Create curved mesh
-cells_per_dimension = (125,)
+cells_per_dimension = (120,)
 coordinates_min = (-1.0,) # minimum coordinate
 coordinates_max = (1.0,)  # maximum coordinate
 mesh = StructuredMesh(cells_per_dimension, coordinates_min, coordinates_max,

diff --git a/examples/structured_1d_dgsem/elixir_euler_sedov.jl b/examples/structured_1d_dgsem/elixir_euler_sedov.jl
@@ -11,15 +11,15 @@ equations = CompressibleEulerEquations1D(1.4)
     initial_condition_sedov_blast_wave(x, t, equations::CompressibleEulerEquations1D)
 
 The Sedov blast wave setup based on Flash
-- http://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel/node184.html#SECTION010114000000000000000
+- https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114000000000000000
 """
 function initial_condition_sedov_blast_wave(x, t, equations::CompressibleEulerEquations1D)
   # Set up polar coordinates
   inicenter = SVector(0.0)
   x_norm = x[1] - inicenter[1]
   r = abs(x_norm)
 
-  # Setup based on http://flash.uchicago.edu/site/flashcode/user_support/flash_ug_devel/node184.html#SECTION010114000000000000000
+  # Setup based on https://flash.rochester.edu/site/flashcode/user_support/flash_ug_devel/node187.html#SECTION010114000000000000000
   r0 = 0.21875 # = 3.5 * smallest dx (for domain length=4 and max-ref=6)
   # r0 = 0.5 # = more reasonable setup
   E = 1.0
@@ -78,8 +78,8 @@ save_solution = SaveSolutionCallback(interval=100,
 
 stepsize_callback = StepsizeCallback(cfl=0.5)
 
-callbacks = CallbackSet(summary_callback, 
-                        analysis_callback, 
+callbacks = CallbackSet(summary_callback,
+                        analysis_callback,
                         alive_callback,
                         save_solution,
                         stepsize_callback)