prepare 0.19.7 release

[SimonDanisch]

• update project tomls
• update NEWS.md
• merge some more PRs!?

[SimonDanisch]

Damn, didn't realize that we now have something of a circular dependency via DelaunayTriangulation depending on MakieCore and Makie depending on DelaunayTriangulation, making it hard to release a breaking MakieCore version....
I guess we can act like it's not breaking, since annyone using MakieCore, will need to install a matching Makie version, that has =MakieCoreVersion...

[MakieBot]

Compile Times benchmark

Note, that these numbers may fluctuate on the CI servers, so take them with a grain of salt. All benchmark results are based on the mean time and negative percent mean faster than the base branch. Note, that GLMakie + WGLMakie run on an emulated GPU, so the runtime benchmark is much slower. Results are from running:

using_time = @ctime using Backend
# Compile time
create_time = @ctime fig = scatter(1:4; color=1:4, colormap=:turbo, markersize=20, visible=true)
display_time = @ctime Makie.colorbuffer(display(fig))
# Runtime
create_time = @benchmark fig = scatter(1:4; color=1:4, colormap=:turbo, markersize=20, visible=true)
display_time = @benchmark Makie.colorbuffer(display(fig))

	using	create	display	create	display
GLMakie	14.62s (14.34, 15.04) 0.28+-	1.35s (1.28, 1.40) 0.04+-	0.95s (0.89, 1.04) 0.05+-	15.54ms (15.31, 15.72) 0.15+-	185.04ms (183.83, 187.19) 1.05+-
master	14.76s (14.33, 15.08) 0.25+-	1.35s (1.29, 1.40) 0.04+-	955.26ms (930.21, 989.13) 21.79+-	15.66ms (15.33, 16.03) 0.24+-	185.47ms (184.38, 187.47) 1.02+-
evaluation	-0.98%, -0.14s invariant (-0.54d, 0.33p, 0.27std)	-0.07%, -0.0s invariant (-0.02d, 0.97p, 0.04std)	-0.98%, -0.01s invariant (-0.26d, 0.64p, 0.03std)	-0.75%, -0.12ms invariant (-0.58d, 0.30p, 0.20std)	-0.24%, -0.44ms invariant (-0.42d, 0.45p, 1.04std)
CairoMakie	12.37s (12.00, 12.78) 0.32+-	1.23s (1.20, 1.27) 0.02+-	267.03ms (262.78, 273.74) 3.58+-	12.03ms (11.89, 12.13) 0.10+-	7.12ms (7.00, 7.24) 0.10+-
master	12.24s (11.84, 12.55) 0.24+-	1.24s (1.19, 1.29) 0.04+-	263.84ms (257.99, 272.49) 5.77+-	12.11ms (11.91, 12.32) 0.16+-	7.16ms (7.08, 7.26) 0.07+-
evaluation	+1.01%, 0.13s invariant (0.45d, 0.42p, 0.28std)	-0.51%, -0.01s invariant (-0.20d, 0.72p, 0.03std)	+1.19%, 3.18ms invariant (0.66d, 0.24p, 4.68std)	-0.72%, -0.09ms invariant (-0.67d, 0.24p, 0.13std)	-0.67%, -0.05ms invariant (-0.56d, 0.32p, 0.08std)
WGLMakie	11.77s (11.75, 11.79) 0.01+-	1.18s (1.14, 1.27) 0.05+-	11.44s (11.36, 11.51) 0.06+-	12.86ms (12.19, 13.59) 0.53+-	1.06s (1.02, 1.10) 0.03+-
master	11.77s (11.72, 11.87) 0.05+-	1.14s (1.11, 1.16) 0.02+-	11.62s (11.48, 11.85) 0.14+-	13.19ms (12.79, 13.63) 0.37+-	1.08s (1.05, 1.13) 0.03+-
evaluation	-0.02%, -0.0s invariant (-0.06d, 0.92p, 0.03std)	+3.43%, 0.04s invariant (1.08d, 0.08p, 0.03std)	-1.52%, -0.17s faster ✓ (-1.64d, 0.01p, 0.10std)	-2.52%, -0.32ms invariant (-0.71d, 0.21p, 0.45std)	-1.74%, -0.02s invariant (-0.69d, 0.22p, 0.03std)

[DanielVandH]

Damn, didn't realize that we now have something of a circular dependency via DelaunayTriangulation depending on MakieCore and Makie depending on DelaunayTriangulation, making it hard to release a breaking MakieCore version....
I guess we can act like it's not breaking, since annyone using MakieCore, will need to install a matching Makie version, that has =MakieCoreVersion...

@SimonDanisch: Is this a significant issue? Currently, MakieCore is a package extension inside DelaunayTriangulation, though that doesn't do anything <1.9. Is there a way I can change some of the code up inside DelaunayTriangulation to help you out here?

[SimonDanisch]

I think now that makie depends on delaunaytriangulation, we could just move the extension code to Makie?

[DanielVandH]

That would be a good idea. Once I have some spare time, I will look into doing that.