Help needed in understanding the "getattr()" function

[numberwhun]

Hello everyone! I am presently going through the "Dive Into Python" tutorial which I obtained from their website. No, this is not for any class, I am self-learning the Python language.

I am in Chapter 4 and reading about "Getting Object References with getattr". First, it say in there that, "you can get a reference to a function without knowing its name until run−time, by using the getattr function". That is the first thing that is a little confusing as it doesn't really elaborate on that too much, that I can tell. What does that mean? I know that you can do:

getattr({}, "clear)
 

and that will tell you that it is the clear function of a dictionary. But, what does that have to do with not knowing the function name till runtime?

Next, if you look at example 4.10, there is a line of code in there as follows:

getattr(li, "append")("Moe")
 

and in the text below the examples section it says:

In case it hasn't sunk in just how incredibly useful this is, try this: the return value of getattr is the method, which you can then call just as if you had said li.append("Moe") directly. But you didn't call the function directly; you specified the function name as a string instead.
 

To tell you the truth, it hasn't really sunk in, even yet. First, it says the return value is the method (append) and that you can then call it as if you had done an:

li.append("Moe")
 

Why not just do the direct call? Why go through all the trouble of the getattr() function?

I guess what I am getting at is, what is the getattr() function really useful for, because right now, I just don't see it? Can someone please explain it in a way that will allow me to see its full usefulness?

Sorry for the ignorance if this tends to be a clouded view!

Regards and Happy New Year!

Jeff

[Motoma]

Hello everyone! I am presently going through the "Dive Into Python" tutorial which I obtained from their website. No, this is not for any class, I am self-learning the Python language.

I am in Chapter 4 and reading about "Getting Object References with getattr". First, it say in there that, "you can get a reference to a function without knowing its name until run−time, by using the getattr function". That is the first thing that is a little confusing as it doesn't really elaborate on that too much, that I can tell. What does that mean? I know that you can do:

getattr({}, "clear)
 

and that will tell you that it is the clear function of a dictionary. But, what does that have to do with not knowing the function name till runtime?

Next, if you look at example 4.10, there is a line of code in there as follows:

getattr(li, "append")("Moe")
 

and in the text below the examples section it says:

In case it hasn't sunk in just how incredibly useful this is, try this: the return value of getattr is the method, which you can then call just as if you had said li.append("Moe") directly. But you didn't call the function directly; you specified the function name as a string instead.
 

To tell you the truth, it hasn't really sunk in, even yet. First, it says the return value is the method (append) and that you can then call it as if you had done an:

li.append("Moe")
 

Why not just do the direct call? Why go through all the trouble of the getattr() function?

I guess what I am getting at is, what is the getattr() function really useful for, because right now, I just don't see it? Can someone please explain it in a way that will allow me to see its full usefulness?

Sorry for the ignorance if this tends to be a clouded view!

Regards and Happy New Year!

Jeff

The cases in the book use strings that were statically typed in. However, those strings could just as easily be taken from user input.

[bvdet]

I have used getattr() a few times. It is useful when you need to get an attribute or method of an object with a string variable or argument. In this example, we pass an 'x', 'y' or 'z' to a function that sorts a list of point objects on the appropriate attribute.

def sortPts(ptlist, key='x', reverse=False):
    '''
    Sort a point list on the 'x', 'y' or 'z' attribute
    '''
    try:
        def cmpItems(a,b):
            if reverse:
                return -(cmp(getattr(a, key.lower()), getattr(b, key.lower())))
            return cmp(getattr(a, key.lower()), getattr(b, key.lower()))
        ptlist.sort(cmpItems)
    except AttributeError, e:
        raise AttributeError, 'Error in sortPoints() - %s' % (e)

[numberwhun]

Ok, I somewhat see it, but the fog is still clouding my understanding. I am going to continue reading and hopefully it will lift with further examples. Thank you to both of you for your help!

Regards and Happy New Year!

Jeff