[Tutor] list of instance objects, access attribute

Lie Ryan lie.1296 at gmail.com
Sat Jun 20 00:30:07 CEST 2009

Alan Gauld wrote:
> "Vincent Davis" <vincent at vincentdavis.net> wrote
> class B():
>  def __init__(self, b1, b2):
>     self.fooa = b1
>     self.foob = b2
> I assume thats what you really meant!
> Ok now I have several instances in a list
> b1 = B(1, 2)
> b2 = B(3, 4)
> b3 = B(9, 10)
> alist = [b1, b2, b3]
>> Lets say for each instance of the class I want to print the value of
>> fooa if it is greater than 5. How do I do this, 
> define a method of the class, say bigprint()
> def bigprint(self, limit=5):
>     if self.fooa > limit: print self.fooa
>> about is how I iterate over the values of fooa. 
> Iterate over the objects and call the method. Make the object do the
> work, your code should not need to know about the internal attributes of
> the object.
> For x in alist:
>      x.bigprint()
>> Is that the right way or is there a better?
>> will this work for methods?
> Methods are how you should do it. Direct access other than for simple
> reading of values is a suspicious design smell. Any processing of or
> rules about the data should be in a method.

Personally, I often thought input/output inside an object is a design
smell (except for debugging), preferring something like this:

class B(object):
    def __init__(...):
    def big(self, limit=5):
        return (self.fooa > limit)

alist = [...]

for y in (x for x in alist if x.big()):
    print y.fooa

although admittably often it could make certain codes more difficult to
write; and in some cases the practical approach would be warranted. This
is especially true as the codebase gets larger.

More information about the Tutor mailing list