The idea of this project is to provide a large subsection of the Gaia DR2 stars (approximately 11M stars). This is not particularly useful for astronomical purposes, however, for general user interface applications, it provides a very useful subset in a relatively easy-to-manage size.

DISCLAIMER: When I say this is not particularly useful, I am begging you not to do science with the output of these tools. I am making a lot of compromises for the sake of performance, ease-of-use and visual appearance!

DISCLAIMER: This repository management for this repo is weird. There will not be history applied to this repo, and instead it will always be a truncated repo.

The actual output of this tool limits the data for the stars to what would be used to display them statically in the night sky. I.e. mas (milli-arc-seconds, 1/mas = parseces), latitude, longitude, color (in approximate Hipparcos colors). NOTE: the hipparcos colors are used because there is much better research and algorithms on that mapping space to RGB.

First, download the database.

wget -l 2 -r http:https://cdn.gea.esac.esa.int/Gaia/gdr2/gaia_source/csv/

Second, build 'process'

make

Third, process the stars. This searches for stars that match the following criteria gmag < 13.5 && gmag > 0. This weeds out a ton of useless celestial cruft.

./process.sh

The output from this first phase is processed.csv this basically includes the lines from the original DR2 database as-is. We can then use this to cross- correlate with our Hipparcos dataset.

Included in this is a Hipparcos dataset parserser. What I originally used to generate a visually appealing starfield. The issue is that it only contained approximately 100,000 stars and a lot of the mas data was very, very bad.

This however does contain the HIP star numbers, which is VERY useful when cross correlating to other databases and especially when correlating to the Stellarium constellationship database.

I have included flat_stars.dat in the hipparcos folder along with the files used to generate that data. The flat_stars.dat file contains star data in the following format:

	RECSTR( flatstartype, typedef struct __attribute__((__packed__)) { \
		uint32_t rascention_bams; \
		int32_t  declination_bams; \
		int16_t parallax_10uas; \
		uint16_t magnitude_mag1000; \
		int16_t bvcolor_mag1000; \
		int16_t vicolor_mag1000; \
		uint32_t hipparcos_id; \
	} flat_star; );

And uses the following algorithm to convert for use in that file:

s.rascention_bams =  ( ( tmp = atof( fields[4] ) )/6.28318530718)*4294967295;
s.declination_bams = ( ( tmp = atof( fields[5] ) )/3.14159265359)*2147483647;
s.parallax_10uas = ( tmp = atof( fields[6] ) ) * 100;
s.magnitude_mag1000 = ( tmp = atof( fields[19] ) ) * 1000;
s.bvcolor_mag1000 = ( tmp = atof( fields[23] ) ) * 1000;
s.vicolor_mag1000 = ( tmp = atof( fields[25] ) ) * 1000;
s.hipparcos_id = hc; // atoi( fields[0] );

This is particularly useful for generally displaying stars. And I have had very good luck with it, so I intend to match this sort of mentality for the rest of this project.

Now that we have flat_stars.dat we can begin the cross-correlation process.