Objects as function argument

Question

I have the following classes in Python:

class String:    
    def clean_string(self, corpus):
        f = open(corpus, 'r')
        raw = f.read().lower()
        f.close()
        raw1 = re.sub(omissis, '', raw) 
        self.string = raw1

    def print_string(self):
        return self.string

class Set:
    def letters_set(self, string):
        self.let_set = set(re.findall(omissis, string))

class Dict:
    def __init__(self, dictionary={}):
        self.dictionary = {} 
        self.string = String()
        self.let_set = Set() 

    def generate_possible_triplets(self, let_set):
        triplet = [(ch1, ch2, ch3) for ch1 in let_set
                                   for ch2 in let_set
                                   for ch3 in let_set]
        [...]

I have a problem with objects as function arguments. Suppose I want to create an instance of the class Set, one of class String and call the method .letters_set(String.string).

What do I have to put as argument inside the parenthesis? the name of the object of class string I will create? a variable referenced to this object? (same applies for the method .generate_possible_triplets in Dict. What form should let_set take?

Your letters_set() just expects a normal python string. Is that your intention or are you trying to integrate it with your custom String class? — jdi
– jdi, Commented Oct 10, 2012 at 20:32
any variable references to an object? just make one and pass it like in foo = 0; func(foo) — unddoch
– unddoch, Commented Oct 10, 2012 at 20:34
On a scale of 1 to "don't ever do that", naming classes after built-in data types is pretty bad. You have UserDict and UserString if you want to extend them - allowing you to add methods while still keeping the basic functionality of the underlying type. — Sean McSomething
– Sean McSomething, Commented Oct 10, 2012 at 20:58
@SeanMcSomething: I agree with the first part about the naming for sure. The UserDict, UserString, etc are kind of old (python 2.2) when there was still old-style classes: The need for this class has been largely supplanted by the ability to subclass directly from dict (a feature that became available starting with Python version 2.2). — jdi
– jdi, Commented Oct 10, 2012 at 21:08

jdi · Accepted Answer · 2012-10-10 20:44:06Z

You would probably just want to make your methods accept instances of your custom classes...

class String:
    ...

class Set:
    def letters_set(self, stringObj):
        # stringObj is a String instance
        self.let_set = set(re.findall(omissis, stringObj.string))

class Dict:
    ...

    def generate_possible_triplets(self, setObj):
        # setObj is a Set instance
        triplet = [(ch1, ch2, ch3) for ch1 in setObj.let_set
                                   for ch2 in setObj.let_set
                                   for ch3 in setObj.let_set]


aString = String()
aSet = Set()
aDict = Dict()

aSet.letters_set(aString)
aDict.generate_possible_triplets(aSet)

The methods can then expect to operate on those classes appropriately to access the attributes. This example is not specifically checking the capabilities of the objects being passed in, but they would obviously raise an exception when you try to access an improper object type that does not have a .string or .let_set attribute.

Collectives™ on Stack Overflow

Objects as function argument

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related