From cd2200eb8196fc3bfd44c63a7c9dcf9cb3b7b594 Mon Sep 17 00:00:00 2001 From: Anshul Singhvi Date: Tue, 30 Jul 2024 10:49:43 +0530 Subject: [PATCH] Get meshimage working with tuple npoints (#267) --- src/mesh_image.jl | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/src/mesh_image.jl b/src/mesh_image.jl index 2aa6c7d6..68e61f84 100644 --- a/src/mesh_image.jl +++ b/src/mesh_image.jl @@ -56,7 +56,7 @@ function Makie.plot!(plot::MeshImage) # A sentinel so we can see if `npoints` changed. If it did, we have to # recreate the mesh, which is slightly time-consuming. If not, though, # then we only have to change the texture (image) which is cheaper. - old_npoints = Ref(0) + old_npoints = Ref{Any}(0) # Handle the transformation onany(plot, plot.converted[1], plot.converted[2], plot.transformation.transform_func, plot.npoints, plot.space, plot.z_level; update=true) do x_in, y_in, tfunc, npoints, space, z_level @@ -79,7 +79,7 @@ function Makie.plot!(plot::MeshImage) ys = LinRange(extrema(y_in)..., last(npoints)) poval = points_observable.val # The array is in a grid, so we have to update them on a grid as well. - for (linear_ind, cartesian_ind) in enumerate(CartesianIndices((npoints, npoints))) + for (linear_ind, cartesian_ind) in enumerate(CartesianIndices((first(npoints), last(npoints)))) p = Point3d(xs[cartesian_ind[1]], ys[cartesian_ind[2]], z_level) poval[linear_ind] = Makie.to_ndim( Point3d,