CroatianMeteorNetwork
Jump to the unmatched star which is furthest from all other matched stars
BackSlash key (above Enter) pans and zooms the view to the star which is furthest away from all the other matched stars.
This gives an even spread of stars across the image and reduces the time spent scrolling.
Still to implement is jump to lowest and highest magnitude unmatched star.
-m followed by the location of the directory holding the mask will prevent stars which are obscured from being drawn, and attempting to match
python -m Utils.SkyFit2 /home/david/platepars/au000u -m /home/david/platepars/au000u
will load the mask.bmp from /home/david/platepars/au000u/
Very cool, this seems like a great feature! Before I merge it, a few suggestions:
- The ] key is not too accessible on some non-english keyboards. I recommend putting a button somewhere in the GUI for this feature, or perhaps give a CTRL + something alternative.
- I suggest that if no path for the -m option is given, the working directory is used to look for the mask.
Could you also check if loading an old state file works? I think every time we add a new argument to the class, we need to modify this function: https://github.com/CroatianMeteorNetwork/RMS/blob/3a4d3058ae05dc30c9e5dd5922d41a8d6d739c0a/Utils/SkyFit2.py#L2194
I belive that the new arguments need to be explicitly defined here.
Have you also checked what happens when there are no stars in the calstars file, i.e. it's empty? This might be relevant when non-RMS data is used.
For masks first choice is to to take the command line option next choice it to use the path in the .config file final choice is to pick up the mask.bmp in the working directory
Not having a CALSTARS results in one being created. An empty file just returns an empty list, so the green circles do not get drawn. The code is directed by the catalog stars, limited by FOV and mask. This does make the coordinate handling awkward, because I need to jump to where they are on the screen with the as yet imperfect platepar; but I don't think I rely on CALSTARS.
I've added in the code to load a mask when resuming from a state, but wouldn't it be better to save the mask in the statefile? This is not what you asked for, so I guess there is a good reason...
This is ready for review - not ready for merging. The gui code needs some work. I need to make a lot of platepars over the next few days, so it will get a good test.