Add LPP code and vetters.py. sm add lpp

[mustaric]

I've added the LPP Code. Let me know if this is ready to merge.

Close #9, close #11 , fix #12

EDIT: Acceptance criteria

• [x] Make test pass.
• [x] Check RTD build to see if API doc renders properly without warnings.

[mustaric]

So I can't get the tests to pass without adding a TCE class. It seems that should be a separate pull request because it is a somewhat larger and separate task...yes?no? To do that I need to merge this branch so I can start developing that code on a new branch with this code available. This is where I really don't understand the git workflow. In the mean time I'll work on the small stuff.

[mustaric]

Ah I see, you looked at main.py. I've stopped using that file and it is out of date. I should have removed the duplicate code and moved the file. ok. I'll make that change.

[pllim]

For a test that you know will fail until more things are fixed in the future, you can try to use https://docs.pytest.org/en/latest/skipping.html#xfail-mark-test-functions-as-expected-to-fail and provide a reason. Thanks!

[pllim]

I pushed this as far as I could, but I don't know how to fix 'Lppdata' object has no attribute 'period', which I think is a real bug. Here is a summary of what I did in my commit:

• Rebased against upstream/master
• PEP 8 fix
• Add docstring and formatting for Sphinx using numpydoc style
• Make the new dependencies optional (import only when needed, not installed when exovetter is installed)
• Add __all__ to guard against namespace in star import
• Vetter is now BaseVetter and abstract
• Removed pseudocode that won't run
• Rearranged imports (order: stdlib, third-party, local) and removed unused import
• Added your new dependencies to setup.cfg and CI
• Deleted commented print statements (they are PY2 style anyway)
• Replaced active print with either warning or exception, as appropriate
• Replace generic Exception with more explicit subclasses
• Validation to raise error before warning
• Tried to guess/fix undefined variables in get_boxmodel
• Ignore some upstream warnings in setup.cfg when running pytest
• Fix bugs exposed by running pytest

I also added acceptance criteria above. Only merge when all boxes are checked.

[mustaric]

It looks like you were using the wrong version of the map file. I fixed that, then uncommented all the parameters that are required to run lpp that are indeed in this version of the map file. I then fixed the test to have more tolerance. It is again passing for me. It doesn't look like the run_one_lpp test is passing above, but it is passing on my machine, so I'm a bit at a loss.

[pllim]

E FileNotFoundError: [Errno 2] No such file or directory: '/Users/smullally/Python_Code/exoplanetvetting/original/dave/lpp/newlpp/map/combMapDR25AugustMapDV_6574.mat'

You need to put the file online somewhere. If it is less than 5 MB, I would say just put it in this repo at this point to get things moving forward.

[pllim]

Also you reintroduced some PEP 8 warnings. FYI.

[pllim]

If you want to put the data file in this package, do the following:

1. Put it in a new exovetter/data/ directory.
2. Add this in setup.cfg:

[options.package_data]
exovetter = data/*

3. To use it in your code (you need to reinstall your local package unless you installed it in editable mode already):

import os

from astropy.utils.data import get_pkg_data_filename

def myfunction(...):
    filename = get_pkg_data_filename(os.path.join('data', 'datafile.mat'))

4. If things work, add the data file and new code into version control.

Please let me know if you need help. Good luck.

[mustaric]

Looks like I had accidentally added one of my local testing files to the github repo...which was using the local copy of the map file. So I removed that and it looks like tests are passing. Next step is to find a way to actually fix the pep8 violations without doing it by hand.