Change name & behavior of RDB object to make less confusing about hydrofunctions

[mroberge]

One solution is to use the namedtuple format.
The functions that return a RDB object would return a tuple that you could unpack like this:
header, table = peaks('01582500')
or, you could just collect the RDB object like this:
my_RDB_namedtuple_obj = peaks('01582500')

The repr() would print out the object like this:

>>> my_RDB_namedtuple_obj
hydroRDB(header='# THIS IS THE USGS HEADER EXCEPT MUCH SHORTER', table=
pandas table goes here)

Right now, I have the RDB object include a header row and the row with the variables' formats. I would get rid of those for now, I think. Just to keep the tuple simple. Otherwise, you have to unpack a four-part-tuple (quadruple?) like this:
a, b, c, d = peaks('01582500')

[mroberge]

I implemented the solution as described above.

a hydroRDB prints itself like a named tuple, and it behaves a little like a named tuple: it will output as its own object or as a tuple of the header and the table. Or you can access these through properties.

[mroberge]

closed with 76f92cf