WIP - replace local_stars.lua and bright_stars.lua with stars from the HYG catalog

[ecraven]

Use the stars in https://github.com/astronexus/HYG-Database.

Conversion done by scripts/convert-hygdata.scm (look at the comments in the file).
The generated file is already included, you don't need to generate it yourself.

This is not meant for inclusion, but to look at the map (F2) on how actual stars would look.

Note: the in-space star background is not real stars (yet).

[impaktor]

"My god! It's full of stars!"

Does it / will it give the same-ish positions as our previous "real" stars?

[ecraven]

Does it / will it give the same-ish positions as our previous "real" stars?

Hard to say exactly, I just checked a few, Sirius has moved a bit and is now right outside the initial jump sphere, but I've tried to align things so that they are kind of in the same direction ;)

[starling13]

Will this commit be merged? This is very interesting step toward realism which not requires code modification and complexity increment.

[Hoimar]

I agree, would love to see this in game!

[impaktor]

I don't have a computer to test this on, sadly, (anyone have a screen shot of the new sector view?)

• I see this adds absolute magnitude as well, nice. I assume that will come in handy when the in-game stars are taken from the galaxy map as well.

• Long term, would we ideally like to have star data in JSON rather than lua? (or some binary format, even?)

• Tinking out loud: Is the solution that custom systems have to remove/comment out the system they add from the catalouge file "good"? Would it be possible to allow custom systems to subtract stars as well? I.e. If you add a custom system script for Sirius, you could in that script define it should remove the closest instance of the system by the same name (or even some other name) and add itself to the galaxy?

Will this commit be merged?

I assume no(?), as this is might be a proof of concept only, but the functionality should/will make it in at some point, I hope.

[impaktor]

@Web-eWorks I suspect we can close this, since #4879 is a more relevant re-implementation of this PR?