Broadcasting in Ska.quatutil.yagzag2radec

[javierggt]

This issue follows issue sot/chandra_aca/issues/102 and it is better described there.

[taldcroft]

I have a fix for this which I believe works. Part of the problem is that we (I?) have chosen the wrong convention for where to place the extra axes in places (e.g. eci2radec). These extra axes were put at the end of the shape, which can make life easier if there is just a single extra axis, but for multiple extra axes I think that breaks down. For Quaternion @javierggt got it right.

def yagzag2radec(yag, zag, q):
    """
    Given ACA Y-ang, Z-ang and pointing quaternion determine RA, Dec. The
    input ``yag`` and ``zag`` values can be 1-d arrays in which case the output
    ``ra`` and ``dec`` will be corresponding arrays of the same length.

    :param yag: ACA Y angle (degrees)
    :param zag: ACA Z angle (degrees)
    :param q: Quaternion
    :rtype: list ra, dec (degrees)
    """
    try:
        one = np.ones(len(yag))
    except TypeError:
        one = 1.0
    d_aca = np.array([one, tan(radians(yag)), tan(radians(zag))])
    d_aca *= 1.0 / np.sum(d_aca**2)
    d_aca_t = np.moveaxis(d_aca, 0, -1)
    eci = np.einsum('...jk,...k->...j', q.transform, d_aca_t)
    eci_t = np.moveaxis(eci, 0, -1)
    return eci2radec(eci_t)

[taldcroft]

One idea that just occurred to me is to basically stop updating Ska.quatutil and copy the functionality we need into Quaternion or chandra_aca.transform, and then make the N-d support work consistently. We could perhaps deprecate Ska.quatutil, but that is not necessarily needed.

[javierggt]

I second this last proposal. I would split it: radec2yagzag and yagzag2radec into chandra_aca.transform, the rest into Quaternion.

[jeanconn]

I may not understand scope, but I was thinking we might want to do something in Ska.quatutil to have those methods just not support a Quaternion with length > 1 and raise an error. That would at least be consistent. If we are moving methods around, we may still want something like that and a deprecation warning.

[javierggt]

well, I think there is not much difference whether we make it work with the vectorized quaternion before or after deprecating and moving.