Currently internally in the code we use angles in degrees in most places and also use this to store inputs and outputs in tables.
I think we should use radians, because that's what all coordinate packages use internally for the calculations, so if we start doing any speed benchmarking, that should be the input / output format of the data we pass in and out. Also changing to radians will get rid of a few lines of conversion code.
@astrofrog - Would you be OK with this change? I could do it now.
Currently internally in the code we use angles in degrees in most places and also use this to store inputs and outputs in tables.
I think we should use radians, because that's what all coordinate packages use internally for the calculations, so if we start doing any speed benchmarking, that should be the input / output format of the data we pass in and out. Also changing to radians will get rid of a few lines of conversion code.
@astrofrog - Would you be OK with this change? I could do it now.
In any case, as mentioned by @astrofrog at https://github.com/astropy/coordinates-benchmark/pull/42#issuecomment-70836638 , we should store the angle units used in the input / output tables, e.g. by using ECSV once it's merged into Astropy.