[python-uk] memoize & ordering of kwargs.items()

[Alexander Harrowell]

On Friday 11 Nov 2011 09:30:23 Jonathan wrote:
> I agree that 'to be sure' is enough of a justification for including 
the 'sorted' call anyway. But it irks me that I can't write a test to 
show it.

This is surely inherent in the fact that python dictionaries are not 
deterministically ordered. The dox don't say that you will always get 
key-val pairs back in *different* orders - just that you cannot rely on 
them being in the *same* order. Equally, you can't assume that they 
*won't*.


Further, the dox also suggest that the behaviour is not truly random, so 
you can't rely on statistical randomness. You just have to avoid the 
whole idea of order when retrieving items from a python dictionary and 
treat it purely as a hash table that supports both a get-item-by-id 
behaviour and also some iterative and setwise methods.


ISTR there are cases where two calls to dict.items() (and dict.keys() 
and dict.values()) *will* get the same order if there was no intervening 
write to the dict, but I suspect it's best to stick to the "right" 
behaviour.

-- 
The only thing worse than e-mail disclaimers...is people who send e-mail 
to lists complaining about them
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-uk/attachments/20111111/fe1d1c8b/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part.
URL: <http://mail.python.org/pipermail/python-uk/attachments/20111111/fe1d1c8b/attachment.pgp>