Better ways for implementing two situations

[MRAB]

On 2019-04-21 19:23, Paulo da Silva wrote:
> Hi all.
> 
> I am looking for improved solutions to these two problems.
> They are to be in a program that deals with big data. So, they need to
> be fast and save memory.
> 
> Problem 1.
> 
> I have a list of objects and want to split it in a list of groups.
> Each group must have all "equal objects" and have more than one object.
> "equal objects" is based on an id we can get from the object.
> Nothing is sorted.
> 
> Here is my implementation:
> 
> splitter={}
> for f in Objs:
> 	splitter.setdefault(f.getId1,[]).append(f)
> groups=[gs for gs in splitter.values() if len(gs)>1]
> 
> 
> Problem 2.
> 
> I know it seems confusing but pls. look at the implementation.
> Here I have a list of lists of objects - groups.
> Now let's say l is a given list from groups, containing objects.
> I want to replace this list l with a new list of sublists where each
> sublist is a list of "equal objects".
> The new l must have more than one list.
> The final result is a list of lists of lists.
> Nothing is sorted.
> 
> Here is my implementation:
> 
> gs=[]
> for g in groups:
> 	splitter={}
> 	for f in g:
> 		splitter.setdefault(f.getId2(),[]).append(f)
> 	gse=list(splitter.values())
> 	if len(gse)>1: gs.append(gse)
> 
> gs is the result.
> 
> Thanks for any comments.
> 
Have you compared the speed with an implementation that uses 
defaultdict? Your code always creates an empty list for each item, even 
though it might not be needed.