Small tweaks to SDSS loaders

[havok2063]

This PR makes some small tweaks to the SDSS data loaders for SDSS-IV APOGEE data. The existing loaders were not correctly auto-identifying the spectrum format, for data from DR17 or from the southern hemisphere LCO 2.5-m. This fixes that.

Additionally it makes a small tweak to the SDSS-IV spec loaders for BOSS, compared to the SDSS-V loaders, to differentiate the two. Previously identify_spectrum_format would return a list ["SDSS-III/IV spec", "SDSS-V spec"] which breaks expectations for identify_spectrum_format. With the tweak, SDSS spec files from SDSS-IV now correctly return the single format "SDSS-III/IV spec", whereas SDSS-V spec data returns "SDSS-V spec".

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 86.89%. Comparing base (e761853) to head (618d1e3).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1217   +/-   ##
=======================================
  Coverage   86.89%   86.89%           
=======================================
  Files          63       63           
  Lines        4572     4572           
=======================================
  Hits         3973     3973           
  Misses        599      599           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[havok2063]

@rosteen small ping on this, just thinking about specutils 2.0, or the next release of 1.x. Is it possible to get this in for then?

[rosteen]

@havok2063 thanks for the ping, I can take a look tomorrow. I'm trying to get a 2.0 release candidate out in the next couple weeks and release by the end of March, is that timeline ok for this? I wasn't planning to do another 1.x but if it's more urgent I could.

[havok2063]

Thanks @rosteen! It's not urgent. That timeline works just fine!

[rosteen]

Thanks! Looks good to me.

[pllim]

I think this broke Jdaviz test. Does this mean Jdaviz test had been wrong all along until now, or this patch is incomplete?

_______________ test_get_valid_format[SDSS-III/IV spec-specviz] ________________

create_fake_fits = <function create_fake_fits.<locals>._create_fits at 0x7f1597272d40>
name = 'SDSS-III/IV spec', expconf = 'specviz'

    @pytest.mark.parametrize('name, expconf',
                             [('MaNGA cube', 'cubeviz'),
                              ('MaNGA rss', 'specviz'),
                              ('JWST s3d', 'cubeviz'),
                              ('JWST s2d', 'specviz2d'),
                              ('JWST x1d', 'specviz'),
                              ('JWST c1d', 'specviz'),
                              ('JWST x1d multi', 'specviz'),
                              ('JWST c1d multi', 'specviz'),
                              ('SDSS-III/IV spec', 'specviz'),
                              ('generic 3d', 'cubeviz'),
                              ('generic 1d', 'specviz')])
    def test_get_valid_format(create_fake_fits, name, expconf):
        """ test correct file format and config is returned """
        filename = create_fake_fits(name)
    
        # for generic files that have no registered specutils format, valid_format will be blank
        name = name if 'generic' not in name else []
    
        valid_format, config = get_valid_format(filename)
>       assert (valid_format, config) == (name, expconf)
E       AssertionError: assert ([], 'specviz2d') == ('SDSS-III/IV...c', 'specviz')
E         
E         At index 0 diff: [] != 'SDSS-III/IV spec'
E         
E         Full diff:
E           (
E         -     'SDSS-III/IV spec',
E         +     [],
E         -     'specviz',
E         +     'specviz2d',
E         ?             ++
E           )

jdaviz/tests/test_data_formats.py:119: AssertionError

[havok2063]

@pllim The test expectation ('SDSS-III/IV spec', 'specviz') is correct. Something in create_fake_fits or get_valid_format might need updating. I'll take a look.

[pllim]

Thanks for the response! @rosteen has a downstream patch at spacetelescope/jdaviz#3489