Delete unused examples + clean up examples#277
Conversation
Removes ten standalone examples that are now covered by tests, doctests, or the surviving end-to-end simulations (download_glorys_data, ecco_inspect_*, generate_bathymetry, generate_surface_fluxes, idealized_single_column, inspect_JRA55_data, inspect_woa_*, mediterranean_simulation_with_ecco_restoring, meridional_heat_transport_ecco). Updates the remaining examples (global_climate_simulation, near_global_ocean_simulation, one_degree_simulation) to: - pass Fields, AbstractOperations, and Observables directly to `heatmap!` (Oceananigans Makie extension fills immersed land cells with `NaN`), - use unicode subscripts (uᵒᶜ/vᵒᶜ, uᵃᵗ/vᵃᵗ, uˢⁱ) for component-specific velocities and temperatures, - bundle multi-component initialization through a single MetadataSet and a `JRA55PrescribedLand` component. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Builds a real `ocean_simulation` on a small LatitudeLongitudeGrid and runs `set!(ocean.model, mset)` with a multi-component ECCO4Monthly MetadataSet, so the routing claim is actually exercised. The expected output reports rounded min/max/mean of `T`, which is stable across runs at this grid resolution. The sea-ice route is described in a comment alongside the call rather than executed, because no existing test/doctest exercises the ECCO sea-ice download path. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
MetadataSet
Codecov Report❌ Patch coverage is
📢 Thoughts on this report? Let us know! |
|
I took the Metatdatset improvements + Metadatase-related fixes in examples into #283. I suggest we merge that and then continue here with visualization improvements, variable changes etc. I did that because building all examples takes ~7hrs so iterating is cumbersome. I also wanna get a docs built on main that succeeds because it's been a while and test whether #237 fixes the issue of tagged releases not having a Docs version; see #60. |
* pull some changes from #277 * Update make.jl
MetadataSet|
Is it true that Makie extension takes observables? |
Can you clarify? Syntax like this: en = @lift e[$n]
hm = heatmap!(axe, en, colorrange = (0, 1e-3), colormap = :solar, nan_color=:lightgray)is used in all the oceananigans, breeze examples. |
Summary
global_climate_simulation.jl,near_global_ocean_simulation.jl, andone_degree_simulation.jl:heatmap!(Oceananigans Makie extension auto-NaNs immersed land cells),uᵒᶜ,uᵃᵗ,uˢⁱ, etc.),one_degree_simulation, drive both ocean and sea-ice init from oneMetadataSetand add aJRA55PrescribedLandcomponent.```juliablock in theset!(model, ::MetadataSet, names)docstring with a runnable```jldoctestthat builds a realocean_simulation, callsset!(ocean.model, mset)with a multi-component ECCO4MonthlyMetadataSet, and reports roundedmin/max/meanof the resulting temperature field. The sea-ice route is mentioned in a comment alongside the call (no existing test/doctest exercises the ECCO sea-ice download path, so we don't run it here).