TypeError from moon_altaz

[cdeil]

With astroplan 0.1, I'm getting this error on one machine (Scientific Linux 6, Python 2.7, in a virtuelenv):

________________________________________________________________________________ test_moon_altaz _________________________________________________________________________________

    @pytest.mark.skipif('not HAS_PYEPHEM')
    def test_moon_altaz():
        time = Time('2012-06-21 03:00:00')
        location = EarthLocation.from_geodetic(-155*u.deg, 19*u.deg, 0*u.m)
        obs = Observer(location=location, pressure=0*u.bar)
>       altaz = obs.moon_altaz(time)

/lfs/l2/hess/software/python/superpy2/lib/python2.7/site-packages/astroplan/tests/test_observer.py:1086: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 

self = <Observer: location (lon, lat, el)=(-155.0 deg, 19.0 deg, 0.0 m),
    timezone=<UTC>,
    pressure=<Quantity 0.0 bar>>
time = <Time object: scale='utc' format='iso' value=2012-06-21 03:00:00.000>

    def moon_altaz(self, time):
        """
            Returns the position of the moon in alt/az.

            TODO: Currently `moon_altaz` uses PyEphem to calculate the position
            of the moon.

            Parameters
            ----------
            time : `~astropy.time.Time` or other (see below)
                This will be passed in as the first argument to
                the `~astropy.time.Time` initializer, so it can be anything that
                `~astropy.time.Time` will accept (including a `~astropy.time.Time`
                object).

            Returns
            -------
            altaz : `~astropy.coordinates.SkyCoord`
                Position of the moon transformed to altitude and azimuth

            Examples
            --------
            Calculate the altitude and azimuth of the moon at Apache Point
            Observatory:

            >>> from astroplan import Observer
            >>> from astropy.time import Time
            >>> apo = Observer.at_site("APO")
            >>> time = Time("2015-08-29 18:35")
            >>> altaz_moon = apo.moon_altaz(time)
            >>> print("alt: {0.alt}, az: {0.az}".format(altaz_moon)) # doctest: +FLOAT_CMP
            alt: -64.1659594407 deg, az: 345.360401117 deg
            """
        if not isinstance(time, Time):
            time = Time(time)

        try:
            import ephem
        except ImportError:
            raise ImportError("The moon_altaz function currently requires "
                              "PyEphem to compute the position of the moon.")

        moon = ephem.Moon()
        obs = ephem.Observer()
>       obs.lat = self.location.latitude.to(u.degree).to_string(sep=':')
E       TypeError: can only update value with String or number

/lfs/l2/hess/software/python/superpy2/lib/python2.7/site-packages/astroplan/observer.py:1502: TypeError
================================================================ 5 failed, 39 passed, 1 skipped in 117.12 seconds ================================================================

https://gist.github.com/cdeil/da030a53521957f74323

I don't understand why this is happening yet, but @bmorris3 if it doesn't make sense to you from the error message, I should be able to debug this quickly on that machine.

[bmorris3]

I don't know what the issue is there.

[eteq]

@cdeil - do you know what version of pyephem this is on? I wonder if that's the source of the problem. I see the following on pyephem v3.7.5.3:

>>> location = EarthLocation.from_geodetic(-155*u.deg, 19*u.deg, 0*u.m)
>>> obs = ephem.Observer()
>>> obs.lat = location.latitude.to(u.degree).to_string(sep=':')

with no error. So I speculate it's some change in pyephem's parsing?

[bmorris3]

@cdeil Would you confirm if this is still an issue with the dev version of astroplan? A lot has changed since this was posted.

[cdeil]

I no longer have access to a similar Python on that machine, so I can't re-test.
Closing this issue now.