Separating elements from a list according to preceding element

[Bruno Desthuilliers]

Rob Cowie a écrit :
> I'm having a bit of trouble with this so any help would be gratefully
> recieved...
> 
> After splitting up a url I have a string of the form
> 'tag1+tag2+tag3-tag4', or '-tag1-tag2' etc. The first tag will only be
> preceeded by an operator if it is a '-', if it is preceded by nothing,
> '+' is to be assumed.
> 
> Using re.split, I can generate a list that looks thus:
> ['tag1', '+', 'tag2', '+', 'tag3', '-', 'tag4']
> 
> I wish to derive two lists - each containing either tags to be
> included, or tags to be excluded. My idea was to take an element,
> examine what element precedes it and accordingly, insert it into the
> relevant list. However, I have not been successful.
> 
> Is there a better way that I have not considered? 

If you're responsible for the original URL, you may consider rewriting 
it this way:
scheme://domain.tld/resource?tag1=1&tag2=1&tag3=1&tag4=0

Else - and after you've finished cursing the guy that came out with such 
an innovative way to use url parameters - I think the first thing to do 
would be to fix the implicit-first-operator-mess, so you have something 
consistent:

if the_list[0] != "-":
   the_list.insert(0, "+")

Then a possible solution could be:

todo = {'+' : [], '-' : []}
for op, tag in zip(the_list[::2], the_list[1::2]):
   todo[op].append(tag)

But there's surely something better...