SubclassingBuiltInTypes¶

Legacy Wiki Page

This page was migrated from the old MoinMoin-based wiki. Information may be outdated or no longer applicable. For current documentation, see python.org.

Discussion¶

I wonder a lot, “How do I subclass built-in types?”

There’s a related question: “How do I figure out if something is sort of like a particular type?”

My goal is to take something, and say, “Does this act pretty much like an integer?” Or “does this act pretty much like a float?” Or “does this act pretty much like a dictionary?”

You could: count up all the behaviors you rely on. Then test if they all exist. If they all exist, consider that item a fit.

But, that seems really hard, if you’ve got a whole class with a ton of methods, all of which use the item. “Counting up” seems pretty hard in that case, and it seems like it would require a lot of discipline.

Is there, then, a set of tests that we can run over an item, to see if it acts “pretty much like” an int, a float, a dict, a string, blah blah blah…? (Or, at least supports the interface for all those things?)

What do people do about this?

Because I read all these things saying, “Don’t use type!” “Don’t use isInstance!” But I don’t see much in the way of what to use.

– LionKimbro 2004-09-12 17:10:47

[lwickjr]: I like to use issubclass. I suppose one could define a module in the Standard Library consisting entirely of empty classes with appropriate names, and inherit from them:

Module interfaces:

...
class dictLike: pass
...

User module:

import interfaces
...
class Shelf(interfaces.dictLike,...):
...

Then other users could write isinstance(Thing, interfaces.dictLike), and be assured that anything that passes the test /promises/ to be “like a dictionary”.

Comments?

For instance, on the int class, there are some gazillion methods:

#!python >>> dir( 42 ) [‘[abs]’, ‘[add]’, ‘[and]’, ‘[class]’, ‘[cmp]’, ‘[coerce]’, ‘[delattr]’, ‘[div]’, ‘[divmod]’, ‘[doc]’, ‘[float]’, ‘[floordiv]’, ‘[getattribute]’, ‘[getnewargs]’, ‘[hash]’, ‘[hex]’, ‘[init]’, ‘[int]’, ‘[invert]’, ‘[long]’,

‘[lshift]’, ‘[mod]’, ‘[mul]’, ‘[neg]’, ‘[new]’, ‘[nonzero]’, ‘[oct]’,

‘[or]’, ‘[pos]’, ‘[pow]’, ‘[radd]’, ‘[rand]’, ‘[rdiv]’, ‘[rdivmod]’, ‘[reduce]’, ‘[reduce_ex]’, ‘[repr]’, ‘[rfloordiv]’, ‘[rlshift]’, ‘[rmod]’, ‘[rmul]’, ‘[ror]’, ‘[rpow]’, ‘[rrshift]’, ‘[rshift]’, ‘[rsub]’, ‘[rtruediv]’, ‘[rxor]’, ‘[setattr]’, ‘[str]’, ‘[sub]’, ‘[truediv]’, ‘[xor]’] >>> }}}

If you want to make something that’s functionally like an int, can you get around having to implement all these methods?

Maybe I have the wrong page title; Maybe this shouldn’t be called “SubclassingBuiltInTypes,” but rather “SimulatingBuiltInTypes.”

– LionKimbro 2004-09-13 00:00:50

SubclassingBuiltInTypes¶

See Also¶

Discussion¶