A Python driver for CQL that adheres to py-dbapi v2
 (PEP249, Python Database API Specification v2.0:  http://www.python.org/dev/peps/pep-0249/).

Standard use:
 >> import cql
 >> con = cql.connect(host, port, keyspace)
 >> cursor = con.cursor()
 >> cursor.execute("CQL QUERY", dict(kw='Foo', kw2='Bar, etc...))

    - cursor.description  # None initially, list of N tuples that represent
                              the N columns in a row after an execute. Only 
                              contains type and name info, not values.
    - cursor.rowcount     # -1 initially, N after an execute
    - cursor.arraysize    # variable size of a fetchmany call
    - cursor.fetchone()   # returns  a single row
    - cursor.fetchmany()  # returns  self.arraysize # of rows
    - cursor.fetchall()   # returns  all rows, don't do this.

 >> cursor.execute("ANOTHER QUERY", **more_kwargs)
 >> for row in cursor:  # Iteration is equivalent to lots of fetchone() calls
 >>     doRowMagic(row)

 >> cursor.close()
 >> con.close()

Query substitution:
 - Use named parameters and a dictionary of names and values. 
    e.g. execute("SELECT * FROM CF WHERE name=:name", {"name": "Foo"})