[Tutor] Little subclass understanding problem

[Michael H. Goldwasser]

On Thursday November 15, 2007, Tom wrote: 

>    I am trying to understand what happens in the following scenario:
>    
>    class Sub_class(Base_class):
>        def __init__(self, data):
>            Base_class.__init__(self, data)
>    
>    as in:
>    
>    # snippet from http://viner.tv/go?set
>    class Set(list):
>        def __init__(self, value = []):
>            list.__init__([])


Tom,

  Indeed, that first line of the Set constructor body is invoking the
  constructor for the base class (list, in this case).  The reason
  that the value parameter is NOT being directly sent to the base
  class is that there is a desire to avoid allowing potentially
  duplicate elements into something that is representing a set. 

  However, it was entirely unnecessary for them to send an empty list
  as a parameter.  It would suffice to have written

             list.__init__()

  It is important to have the call to the base class initializer
  because we need to allow for the internal state of the underlying
  list to be properly intitialized. 

  Please note that the original source for this from viner.tv
  has an important fourth line:

    class Set(list):
        def __init__(self, value = []):
            list.__init__([])
            self.concat(value)              # copies mutable default

  That fourth line uses a custom method defined later to insert
  designated values into the set while making sure to avoid
  duplicates.

With regard,
Michael