views:

102

answers:

2

Hi, I have an uniform list of objects in python:

class myClass(object):
    def __init__(self, attr):
        self.attr = attr
        self.other = None

objs = [myClass (i) for i in range(10)]

Now I want to extract a list with some attribute of that class (let's say attr), in order to pass it so some function (for plotting that data for example)

What is the pythonic way of doing it,

attr=[o.attr for o in objsm]

?

Maybe derive list and add a method to it, so I can use some idiom like

objs.getattribute("attr")

?

+3  A: 

attrs = [o.attr for o in objs] was the right code for making a list like the one you describe. Don't try to subclass list for this. Is there something you did not like about that snippet?

Mike Graham
Only that it's too verbose and I'm going to be using it often, so a shortcut would be handy.numpy's structured arrays arr['colname'] or recarrays arr.colname, already provide something similar, but in this case I'm using objects that hold named data (and some functions).
franjesus
A: 

You can also write:

attr=(o.attr for o in objsm)

This way you get a generator that conserves memory. For more benefits look at Generator Expressions.

aeby