local variable 'tsOut' referenced before assignment when using own arrays for lightcurve data

[warrickball]

The following snippet

import numpy as np
import pbjam

t = np.arange(1000.)
y = np.sin(t/10.)
s = pbjam.session(ID='tsOut_error', timeseries=[t, y],
                  numax=(240.0, 10.0), dnu=(20.0, 0.2),
                  teff=(5000, 100), bp_rp=(1.0, 0.02))
s(make_plots=True)

produces the error

Checkpoint exception raise
/home/wball/.local/lib/python3.7/site-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.
  import pandas.util.testing as tm
Traceback (most recent call last):
  File "tsOut_error.py", line 8, in <module>
    teff=(5000, 100), bp_rp=(1.0, 0.02))
  File "/home/wball/pypi/PBjam/pbjam/session.py", line 579, in __init__
    lkwargs)
  File "/home/wball/pypi/PBjam/pbjam/session.py", line 356, in _lc_to_lk
    if tsOut:
UnboundLocalError: local variable 'tsOut' referenced before assignment

It looks like the problem is that this option hasn't been implemented in session._lc_to_lk. The main block looks like this (starting at line 332):

    if isinstance(tsIn, str):
        try:
            t, d = np.genfromtxt(tsIn, usecols=(0, 1), delimiter = ',').T 
        except:
            try:
                t, d = np.genfromtxt(tsIn, usecols=(0, 1), delimiter = ' ').T
            except:
                raise IOError('Failed to read the provided ascii files. Please check that they have the required 2-column format, and they are use either comma or white-space delimiters.')
        
        d += tinyoffset
        
        tsOut = lk.LightCurve(time=t, flux=d, targetid=ID)
        
    elif not tsIn:
        if specIn:
            pass
        else:
            tsOut = _query_lightkurve(ID, download_dir, use_cached, lkwargs)

    elif tsIn.__module__ == lk.lightcurve.__name__:
        pass
    else:
        raise TypeError("Can't handle this type of time series object")

    if tsOut:
        _sort_lc(tsOut)

    return tsOut

I checked the type of tsIn in my snippet and it should be caught by tsIn.__module__ == lk.lightcurve.__name__ but that just leads to pass, so tsOut is never set. Because it isn't implemented, I thought it would've been left for the else clause to raise a TypeError but maybe someone intended to work on it.

I've honestly gone cross-eyed with all the wrapping in DataFrame and LightCurve objects so I'm not sure what's meant to happen. Maybe just return tsIn?

[nielsenmb]

Yup, that elif line just asks if what was given is a lightkurve time series. In which case it should have just been returned. It should be passed to _sort_lc(tsOut) first though, as it does some sigma clipping and flattening if I remember correctly, so replacing pass with tsIn=tsOut should fix this issue.

[warrickball]

tsOut = tsIn has worked for me.