[Python-3000] Rounding in Py3k (original) (raw)

Ron Adam rrr at ronadam.com
Fri Aug 4 12:46:42 CEST 2006

Greg Ewing wrote:

Ron Adam wrote:

I'm still not sure why "round" should be preferred in place of "round" as a method name. There isn't an operator associated to rounding so wouldn't the method name not have underscores? I was thinking there would be functions such as round(), trunc(), etc. that use round to do their work. That's why I called it a protocol and not just a method. -- Greg

I understood your point. :-)

If you look at the methods in int, long, and float, there are no methods that do not have double underscores. While there are many that don't in unicode and string. There also are many methods in Decimal that do not use the double underscore naming convention.

I am just curious why not in general for the builtin numeric types.

The style guide says...

- doubleleadingandtrailingunderscore: "magic" objects or attributes that live in user-controlled namespaces. E.g. init, import or file. Never invent such names; only use them as documented.

So would round interact with the interpreter in some "magic" way? I take "magic" to mean the interpreter calls the method directly at times without having python coded instructions to do so. Such as when we create an object from a class and init gets called by the interpreter directly. The same goes for methods like add and repr, etc...

But that doesn't explain why int, long, and float, don't have other non-magic methods.

I'm not attempting taking sides for or against either way, I just want to understand the reasons as it seems like by knowing that, the correct way to do it would be clear, instead of trying to wag the dog by the tail if you know what I mean.

Cheers, Ron

More information about the Python-3000 mailing list