Would there be support for a more general cmp/__cmp__

Antoon Pardon apardon at forel.vub.ac.be
Thu Oct 20 13:13:28 CEST 2005

Op 2005-10-20, Steve Holden schreef <steve at holdenweb.com>:
> Antoon Pardon wrote:
>> I was wondering how people would feel if the cmp function and
>> the __cmp__ method would be a bit more generalised.
>> The problem now is that the cmp protocol has no way to
>> indicate two objects are incomparable, they are not
>> equal but neither is one less or greater than the other.
>> So I thought that either cmp could return None in this
>> case or throw a specific exception. People writing a
>> __cmp__ method could do the same.
> The current behaviour is, of course, by design: """The operators <, >, 
>==, >=, <=, and != compare the values of two objects. The objects need 
> not have the same type. If both are numbers, they are converted to a 
> common type. Otherwise, objects of different types always compare 
> unequal, and are ordered consistently but arbitrarily."""
> Personally I'm still not convinced that your requirement reflects a 
> substantial use case (but then I'm getting used to that ;-). Just 
> because an ordering is partial that doesn't mean that two instances of a 
> class shouldn't be compared.

I'm not suggesting that they shouldn't be compared.

> What would you have Python do when the programmer tries to perform an 
> invalid comparison (i.e. what are the exact semantics imposed when 
> __cmp__() returns None/raises an exception)?

My appologies, I should have been more complete.

What I want is a way to say that all of the following are False:

  a < b, a <= b, a == b, a > b, a >= b

and that only the following is True:

  a != b

So if a coder writes one of the comparisons and as a result python
calls the __cmp__ method on one of the terms which would return
either None or raise an exceptions I would like it to reflect
the above behaviour.

Antoon Pardon

More information about the Python-list mailing list