Add API and attr support for include/exclude stars for guide

[taldcroft]

This makes it possible to specify a list of stars to specifically include or exclude for both acq and guide catalogs. This changes the get_aca_catalog API to have separate kwargs for acq and guide.

• The include/exclude functionality is implemented and working for acq catalog selection.
• For guide it is only a stub but there are minimal tests showing that it does set the appropriate GuideCatalog include_ids and exclude_ids attributes.

[taldcroft]

I really should define a descriptor to handle these repetitive definitions.

[jeanconn]

The '_acq' seems a little odd for the halfw (because we don't handle halfw for guide or fid), but I suppose it is consistent.

[jeanconn]

What is the current acq behavior if you try to include a star that is not allowed? Warn and proceed if it is on the CCD. Warn and ignore if it isn't?

[taldcroft]

It needs to be a star in the ACA FOV (as defined by get_stars()). If not proseco raises an exception.

proseco/proseco/acq.py

Line 432 in eb9ec2e

for include_id in self.include_ids:

[jeanconn]

It looks from the next line like it needs to be a valid candidate and not just on FOV (meaning the user couldn't say, override COLOR1!=0.7 or pick ASPQ=41). I'm OK with that, just keeping in mind use cases for an override from include_ids.

[taldcroft]

This function is making sure the included star is in the candidate acq star table. The next line checks if it is already a candidate acq star, in which case no action is required. If not already a candidate, then it ensures that the star is in stars, which is an unfiltered list of stars in the ACA FOV.

[jeanconn]

Ah. Thanks!

[taldcroft]

And just to really convince myself. 😄

In [19]: aca = get_aca_catalog(8008, include_ids_acq=[31072944], include_halfws_acq=[120], raise_exc=True)

In [20]: aca
Out[20]: 
<ACATable length=12>
 slot  idx     id    type  sz   p_acq    mag    maxmag   yang     zang    dim   res  halfw
int64 int64  int64   str3 str3 float64 float64 float64 float64  float64  int64 int64 int64
----- ----- -------- ---- ---- ------- ------- ------- -------- -------- ----- ----- -----
    0     1        1  FID  8x8   0.000    7.00    8.00   919.83  -844.17     1     1    25
    1     2        5  FID  8x8   0.000    7.00    8.00 -1828.23  1053.81     1     1    25
    2     3        6  FID  8x8   0.000    7.00    8.00   385.81  1697.85     1     1    25
    3     4 31075128  BOT  6x6   0.974    9.35   10.85  -318.30  1202.29    20     1   160
    4     5 31075368  BOT  6x6   0.985    9.13   10.63    53.94   754.67    20     1   160
    5     6 31983336  BOT  6x6   0.985    8.64   10.14   890.59 -1600.44    20     1   160
    6     7 32374896  BOT  6x6   0.983    9.17   10.67  2022.93 -2021.75    20     1   160
    7     8 31463496  GUI  6x6   0.000    9.46   10.96  2026.70  1399.47     1     1    25
    7     9 31076560  ACQ  6x6   0.940    9.70   11.20  -932.85  -354.64    20     1   160
    0    10 32375384  ACQ  6x6   0.927    9.79   11.29  1612.15  -428.32    20     1   160
    1    11 31982136  ACQ  6x6   0.970   10.19   11.69   561.96  -186.48    20     1    60
    2    12 31072944  ACQ  6x6   0.000   12.40   13.90 -2572.86  1225.46    20     1   120

In [21]: aca.acqs.stars.get_id(31072944)['ASPQ1']
Out[21]: 293

[taldcroft]

Any objection to merging? This passes tests on Mac, Ska3-flight (64) and Matlab32.

[taldcroft]

I wish I could test on Windows, but so it goes. I'm planning to tag at 4.1 after merging this, noting that there have been non-trivial changes since 4.0.