Fix MAG_ACA_ERR units for guide selection

[jeanconn]

Fix MAG_ACA_ERR units for guide selection

I considered adding a new column to the candidate list, but then just used the 0.01 conversion in the two new places that this is used thanks to #105 .

This fix and mag err use in general probably needs more review and better tests, but this might be enough for an updated release candidate for today.

Closes #120

[taldcroft]

Did you check the #120 obsid?

[jeanconn]

Yes, I commented over there on #120

[taldcroft]

👍 with #125 in place.

[jeanconn]

By "with #125 in place" did you mean with the issue in the list or with the test made?

[taldcroft]

With the issue in the list.

[taldcroft]

This is failing for me. Did it pass tests for you?

================================================ FAILURES ================================================
_______________________________________ test_get_aca_catalog_20259 _______________________________________

    @pytest.mark.skipif('not HAS_SC_ARCHIVE', reason='Test requires starcheck archive')
    def test_get_aca_catalog_20259():
        """
        Test obsid 20259 which has two spoiled fids: HRC-2 is yellow and HRC-4 is red.
        Expectation is to choose fids 1, 2, 3 (not 4).
        """
        # Force not using a bright star so there is a GUI-only (not BOT) star
        aca = get_aca_catalog(20259, raise_exc=True)
        exp = ['slot idx     id    type  sz   yang     zang   dim res halfw',
               '---- --- --------- ---- --- -------- -------- --- --- -----',
               '   0   1         1  FID 8x8 -1175.03  -468.23   1   1    25',
               '   1   2         2  FID 8x8  1224.70  -460.93   1   1    25',
               '   2   3         3  FID 8x8 -1177.69   561.30   1   1    25',
               '   3   4 896009152  BOT 6x6  1693.39   217.92  20   1   160',
               '   4   5 897712576  BOT 6x6 -1099.95  2140.23  20   1   160',
               '   5   6 897717296  BOT 6x6   932.58  1227.48  20   1   160',
               '   6   7 896013056  BOT 6x6  1547.25 -2455.12  20   1   160',
               '   7   8 896008536  GUI 6x6  1705.59   158.91   1   1    25',
               '   7   9 896011576  ACQ 6x6   810.99   -69.21  20   1   160',
               '   0  10 896009240  ACQ 6x6  -911.41   402.62  20   1   160',
               '   1  11 897192352  ACQ 6x6 -2110.43  2005.21  20   1   100',
               '   2  12 897719248  ACQ 6x6  1650.09  2019.62  20   1    80']
        repr(aca)  # Apply default formats
>       assert aca[TEST_COLS].pformat(max_width=-1) == exp
E       AssertionError: assert ['slot idx   ...1   160', ...] == ['slot idx    ...1   160', ...]
E         At index 9 diff: '   7   8 896009240  BOT 6x6  -911.41   402.62  20   1   160' != '   7   8 896008536  GUI 6x6  1705.59   158.91   1   1    25'
E         Right contains more items, first extra item: '   2  12 897719248  ACQ 6x6  1650.09  2019.62  20   1    80'
E         Use -v to get the full diff

proseco/tests/test_catalog.py:80: AssertionError

[jeanconn]

I did run the tests, but I guess I've got to be better about having the terminal print branch and commit. I must have retested in master or the like instead of properly testing in my branch.

[taldcroft]

OK, I'll make a fix.

[jeanconn]

Thanks. And I just figured out what I did. I checked out master to reproduce the error in #120 and then forgot that I was in the wrong branch for my run-for-record test. So less a general process fail and more a specific process fail related to trying to see an error in another branch.

[taldcroft]

I've been more in the habit lately of running all tests before making a commit, esp. for something that I expect to pass (not a WIP). And likewise I'm trying to run all tests immediately before pushing to GitHub. This is part of why I want to keep our unit tests reasonably fast.