Fixed a ungettext() call to prevent spurious string extraction by xgettext.

def model_ngettext(obj, n=None):

    """

    Return the appropriate `verbose_name` or `verbose_name_plural` for `obj`

    depending on the count `n`.

    Return the appropriate `verbose_name` or `verbose_name_plural` value for

    `obj` depending on the count `n`.

    `obj` may be a `Model` instance, `Model` subclass, or `QuerySet` instance.

    If `obj` is a `QuerySet` instance, `n` is optional and the length of the

            n = obj.count()

        obj = obj.model

    d = model_format_dict(obj)

    return ungettext(d['verbose_name'], d['verbose_name_plural'], n or 0)

    singular, plural = d["verbose_name"], d["verbose_name_plural"]

    return ungettext(singular, plural, n or 0)