I want a counter to keep count how many iterations in total I have done. I do:
counter = 0;
for i, item in enumerate(items):
for j, anotheritem in enumerate(anotheritems):
counter += 1;
But I can't help asking is there a more pythonic way?
It's simple and readable, that's about as Pythonic as it gets (apart from the semi-colons.)
My only advice would be to give i and j more descriptive names if you start putting more code into those loops.
I would say the most pythonic way is to record i and j rather than count, since the tuple (i,j) has much more meaning than count - you can calculate count from (i,j)!
for i, item in enumerate(items):
for j, anotheritem in enumerate(anotheritems):
where_i_am = (i,j)
# count = i*len(anotheritems) + j + 1
total_iterations = len(items) * len(anotheritems) # equivalently (i+1) * (j+1)
for counter,ijpair in enumerate(itertools.product(enumerate(items),enumerate(other))):
i,j = ijpair
//... do stuff
As Codemonkey points out, we're not increasing readability here, and for the explicit use stated, it's not really an improvement. However, the enumerate and itertools.product expressions are all generators/iterators, so they can be passed to other functions like map to be used there. So you could filter for wherever the item and otheritem were the same value:
allelements = itertools.product(items, others)
sameys = itertools.ifilter(lambda a: a[0]==a[1], allelements)
And now sameys has exactly the same structure as allelements. The point is that the whole iterable set becomes an iterator expression, rather than being a loop construct, which can be more easily passed around.
counter, the iterators can be infinite, it conforms to 'flat is better than nested' Zen, itertools is one of the batteries included in Python (product is v2.6). But yes, it is less 'simple' to read.
item and anotheritem?