Are decorators really that different from metaclasses...

Paul Morrow pm_mon at
Sun Aug 29 18:32:05 CEST 2004

Alex Martelli wrote:
> Paul Morrow <pm_mon at> wrote:
>    ...
>>Cool!  Thanks!  Now I need to ponder why a function's docstring needs a
>>different implementation than a class's docstring (i.e. why not just 
>>make it a straigtforward attribute of the function object).
> Historically, functions had docstrings before they had straightforward
> attributes.  Maybe it could be that.
> Alex

I have stared at your reply for three or four minutes now.  On the 
surface, it seems that you are being genuinely helpful, and so I want to 
reply "Thanks. Of course.  That makes perfect sense."

But then a lot of smart people say things that (on the surface) appear 
honestly helpful but in reality are suggestions that the reader do more 
research before asking questions that have such an obvious answer.

I know that you're a smart person Mr. Martelli.  So I flipped a coin to 
decide which intreprtation of your reply to use, which came up tails, 
indicating the 2nd one.

Bahh. I prefer the superficial one (plz correct me if I'm wrong).

Thanks.  That does make good sense.  They've had no reason to simplify 
the implementation for function docstrings, so "why fix it if it ain't 


More information about the Python-list mailing list