Does colormaps.py require a \n at the end?

Does colormaps.py require a \n at the end?

Yes.

And while everyone's in the nitpicky mode here: Would it make sense to put turbo directly below jet in those lists to allow for a better comparisson?

Good idea, did that and added the \n.

I guess the test are the docs: https://25132-1385122-gh.circle-artifacts.com/0/home/circleci/project/doc/build/html/gallery/color/colormap_reference.html#sphx-glr-gallery-color-colormap-reference-py

Actually, how does turbo compare with Peter Kovesi's rainbow, available in colorcet? see https://colorcet.pyviz.org/

The colorcet rainbow doesn't centre on green, which is one of the things that makes jet/turbo useful.

Just to note something said in a discussion thread:

So since changing the broader profile is outside of scope of 'tweaking', I'm just going to call this one done and leave improvements for a future date :)

As a policy, Matplotlib does not make changes to released colormaps. While you would have until the v3.3 release to make any changes you feel is necessary, this is really why I have always felt that a publication requirement was important for colormaps. The viridis colormap went through a lot of vetting and modifications before it was even named (it was originally called "option D", if I remember correctly). The Brewer colormaps were all published elsewhere before getting included in matplotlib. A few others were all tweaked and modified as part of a peer review process prior to publication that finalized them.

If you want to do major changes to the colormap, then I would suggest doing so and going through the publication route again. Or, perhaps consider calling such a modified colormap a new colormap with a new name (and again, go through a publication route). But don't expect to be able to make any further changes to Turbo in this package once it is in a matplotlib release.

Understood. I think we're actually saying the same thing, because I agree. I basically was trying to say that small tweaks to address people's feedback didn't work, and big changes would mean essentially a new colormap, new publication etc, so let's add turbo as is with it's strengths+weaknesses, and not confuse the matters. If someone comes up with (substantial) improvements to this, it'll likely look different enough for a new publication, discussion, and so on.

re. the above two comments, I'll mark this as release critical so we don't forget it for 3.3, but leave it to stew in case any changes want to be made in the next ~6 months.

@mikhailov-work Have there been any updates to turbo? I am going to rebase and force-push so it will merge cleanly. We are getting close to the 3.3 release so we are getting close to the point of no return ;)

Please let us know if there are any issues, the final release isn't out yet.

does this need a what's new entry?

Ah, yes, that sounds like a good idea.