Silly function call lookup stuff?

[Lucas Lemmens]

On Tue, 27 Sep 2005 13:56:53 -0700, Michael Spencer wrote:

> Lucas Lemmens wrote:
>> Dear pythonians,
>> 
>> I've been reading/thinking about the famous function call speedup trick
>> where you use a function in the local context to represent a "remoter"
>> function to speed up the 'function lookup'.
>> 
>> "This is especially usefull in a loop where you call the function a
>> zillion time" they say.
>> 
>> I think this is very odd behavior.
>> 
>> Why isn't the result of the first function-lookup cached so that
>> following function calls don't need to do the function-lookup at all?
>> 
> I guess because the function name may be re-bound between loop iterations.
>  Are there good applications of this?  I don't know.

Yuk I'd hate that. I think it would be extremely rare.

Would the myLocalFunc = hisRemoteFunc optimization break in such a case?

If not then why not auto-add a local hisRemoteFunc that points to the
remote hisRemoteFunc to the local context when hisRemoteFunc
is executed for the first time.

> 
>> And if the context changes (an import-statement say) reset the cached
>> 'function-lookups'.
> 
> In general an object doesn't know what names are bound to it and there are
> many ways besides an import statement of binding/re-binding, so "if the
> context changes" is easier said than done.
> 

My guess (but I'm not a python programmer) is that context changes would
be the odd case.

So optimizing for not having them ...

> 
>> This way any function would only need to be looked up once.
>> 
>> L.
>> 
> Would you apply this optimization to all lookups in outer scopes, or just
> callables?  Why? ;-)

Hmmm callables have probably the highest chance of being recalled.

> 
> Michael