Remove `ndims` where not necessary #2081

DanielDoehring · 2024-09-17T08:40:49Z

Closes #2080

github-actions · 2024-09-17T08:41:04Z

Review checklist

This checklist is meant to assist creators of PRs (to let them know what reviewers will typically look for) and reviewers (to guide them in a structured review process). Items do not need to be checked explicitly for a PR to be eligible for merging.

Purpose and scope

The PR has a single goal that is clear from the PR title and/or description.
All code changes represent a single set of modifications that logically belong together.
No more than 500 lines of code are changed or there is no obvious way to split the PR into multiple PRs.

Code quality

The code can be understood easily.
Newly introduced names for variables etc. are self-descriptive and consistent with existing naming conventions.
There are no redundancies that can be removed by simple modularization/refactoring.
There are no leftover debug statements or commented code sections.
The code adheres to our conventions and style guide, and to the Julia guidelines.

Documentation

New functions and types are documented with a docstring or top-level comment.
Relevant publications are referenced in docstrings (see example for formatting).
Inline comments are used to document longer or unusual code sections.
Comments describe intent ("why?") and not just functionality ("what?").
If the PR introduces a significant change or new feature, it is documented in NEWS.md with its PR number.

Testing

The PR passes all tests.
New or modified lines of code are covered by tests.
New or modified tests run in less then 10 seconds.

Performance

There are no type instabilities or memory allocations in performance-critical parts.
If the PR intent is to improve performance, before/after time measurements are posted in the PR.

Verification

The correctness of the code was verified using appropriate tests.
If new equations/methods are added, a convergence test has been run and the results
are posted in the PR.

Created with ❤️ by the Trixi.jl community.

codecov · 2024-09-17T09:21:46Z

Codecov Report

Attention: Patch coverage is 81.25000% with 3 lines in your changes missing coverage. Please review.

Project coverage is 96.34%. Comparing base (288d41d) to head (f17e398).
Report is 4 commits behind head on main.

Files with missing lines	Patch %	Lines
src/callbacks_step/amr_dg2d.jl	71.43%	2 Missing ⚠️
src/callbacks_step/amr_dg3d.jl	80.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #2081      +/-   ##
==========================================
+ Coverage   96.32%   96.34%   +0.02%     
==========================================
  Files         470      470              
  Lines       37485    37494       +9     
==========================================
+ Hits        36106    36121      +15     
+ Misses       1379     1373       -6

Flag	Coverage Δ
unittests	`96.34% <81.25%> (+0.02%)`	⬆️

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.

ranocha · 2024-09-17T14:54:51Z

src/callbacks_step/amr_dg2d.jl

        resize!(u_ode,
-                nvariables(equations) * nnodes(dg)^ndims(mesh) * nelements(dg, cache))
+                nvariables(equations) * nnodes(dg)^2 * nelements(dg, cache))
        u = wrap_array(u_ode, mesh, equations, dg, cache)


I think parts like these become much harder to understand with this PR. What do you think, @sloede?

Yes, I have to agree with Hendrik.

The advantage of explicitly adding ndims(mesh) is that it makes the factor self-explanatory. Also in the other parts of the code, one avoids having to explain why a factor is what it is and how it was computed, or even how to change it when switching to another dimension.

I see the argument for more "elegant" code (avoiding the impression of having a varying dimension). However, I've seen too many codes where "magic" constants appear out of nowhere, and everyone reading it after 3 years has to re-understand what the intent/thought process of the original author was.

IMHO, the current version is concise and self-explanatory in terms of what is being done, thus I'd prefer to keep it as it is.

To throw in my two cents. I want to keep the ndims calls in the code. I think it is much more informative to have terms like 2^ndims(mesh) - 1 rather than values like 3 and 7. This is coming from the perspective of someone who recently became very familiar with the AMR implementation as I debugged and tracked down the loss of conservation. I think the current code is perfectly understandable.

Ok, I will close this PR then. Opinions about the status of the issue?

I would close it in light of the comments above

src/callbacks_step/amr_dg2d.jl

src/callbacks_step/amr_dg1d.jl

src/callbacks_step/amr_dg3d.jl

src/callbacks_step/amr_dg2d.jl

DanielDoehring · 2024-09-17T15:01:07Z

src/callbacks_step/amr_dg2d.jl

@@ -330,7 +330,7 @@ function coarsen!(u_ode::AbstractVector, adaptor,
                # If an element is to be removed, sanity check if the following elements
                # are also marked - otherwise there would be an error in the way the
                # cells/elements are sorted
-                @assert all(to_be_removed[old_element_id:(old_element_id + 2^ndims(mesh) - 1)]) "bad cell/element order"
+                @assert all(to_be_removed[old_element_id:(old_element_id + 2^2 - 1)]) "bad cell/element order"


One could move skip = 3 already up here

DanielDoehring · 2024-09-17T15:01:23Z

src/callbacks_step/amr_dg3d.jl

@@ -250,7 +250,7 @@ function coarsen!(u_ode::AbstractVector, adaptor,
                # If an element is to be removed, sanity check if the following elements
                # are also marked - otherwise there would be an error in the way the
                # cells/elements are sorted
-                @assert all(to_be_removed[old_element_id:(old_element_id + 2^ndims(mesh) - 1)]) "bad cell/element order"
+                @assert all(to_be_removed[old_element_id:(old_element_id + 2^3 - 1)]) "bad cell/element order"


One could move skip = 7 already up here

DanielDoehring · 2024-09-17T15:05:13Z

src/callbacks_step/amr_dg1d.jl

@@ -187,13 +187,15 @@ function coarsen!(u_ode::AbstractVector, adaptor, mesh::TreeMesh{1},
                # If an element is to be removed, sanity check if the following elements
                # are also marked - otherwise there would be an error in the way the
                # cells/elements are sorted
-                @assert all(to_be_removed[old_element_id:(old_element_id + 2^ndims(mesh) - 1)]) "bad cell/element order"
+                @assert all(to_be_removed[old_element_id:(old_element_id + 2 - 1)]) "bad cell/element order"


One could move skip = 1 already up here

DanielDoehring · 2024-09-24T06:58:46Z

See discussion

Remove ndims where not necessary

f7a7128

DanielDoehring added discussion low-priority consistency Make Michael happy refactoring Refactoring code without functional changes labels Sep 17, 2024

remove eval

70bc058

ranocha reviewed Sep 17, 2024

View reviewed changes