For loop with changing range

Question

I got a little question: I got 2 lists with equal lengths.

As you can see below in my code, if appropriate circumstances take place, I delete some of the elements from the list. I have to do my work very carefully so my question is - does the for loop check by every iteration if len(whole_st_gen) is changing ?? Won't it skip some items??

whole_st_gen = []
whole_end_gen = [] // the length of both of them is 38273
if_merge = 0
here_merge = 0
deleting = False

for x in range (0, len(whole_st_gen)):
    if_merge = x
    if x == len(whole_st_gen)-1:
        break
    for y in range (x+1, len(whole_st_gen)):
        if whole_end_gen[x]>whole_end_gen[y]:
            deleting = True
            here_merge = y
            continue
        else:
            break
    if deleting == True:
        deleting = False
        del whole_st_gen[x:here_merge]
        del whole_end_gen[x:here_merge]
    if_merge = 0
    here_merge = 0

print len(whole_st_gen) # here's length is 1852, so i think it could work properly, just want to be sure!
sys.exit()

Answer 1

No , when you are using range() method, it does not check the length of the array in each iteration. Especially in Python 2.x , range() returns a list , which is what you iterate over, and this list is created at the start of the loop , it does not get recalculated in each iteration. So there can be multiple issues you can run into when using above method.

One being that you can skip some elements, since if you delete an element from the list, the indices of the list are rearranged to make a contigous sequence, so you would end up missing some elements.

Secondly, you can end up getitng IndexError , Simple example of that -

>>> l = [1,2,3,4]
>>> for i in range(0,len(l)):
...     del l[i]
...
Traceback (most recent call last):
  File "<stdin>", line 2, in <module>
IndexError: list assignment index out of range

This is because like I said in starting range() computes the complete range (even xrange() in Python 2.x) at the start itself. This does not happen for you because of -

if x == len(whole_st_gen)-1:
    break

Like said in the comments an easier way for you to go would be to use while loop , which does not increment when deleting the item. Example -

if_merge = 0
here_merge = 0
deleting = False
x = 0
while x < len(whole_st_gen):
    for y in range (x+1, len(whole_st_gen)):
        if whole_end_gen[x]>whole_end_gen[y]:
            deleting = True
            here_merge = y
        else:
            break
    if deleting == True:
        deleting = False
        del whole_st_gen[x:here_merge]
        del whole_end_gen[x:here_merge]
    else:
        x += 1
    here_merge = 0

print len(whole_st_gen)
sys.exit()

Answer 2

In the for statement the expression list (after the in) is evaluated first to get an iterator (or iterable) which is then used for the iteration, the expression list is not evaluated again in each iteration.

In your case you call the range function to retrieve an iterator (in python3) or iterable (in python2 in the form of a list). In both cases the iterator/iterable will yield len(whole_st_gen) (as it evaluates when you reach the for statement from above) numbers (ranging from zero). If you alter the whole_st_len in the loop it will not affect the iterator and it will yield the same numbers anyway (leading to you skipping numbers, and eventually get numbers that are beyond the indices for whole_st_gen).

As opposed to the situation where you iterate over the container itself the behaviour when you do this is fully specified. You may delete and insert elements in the container if you are careful.

Answer 3

Here is a straight answer to your question, you should not delete items on a list while iterating on the same list, it yields unexpected results to you as some items will be skipped. Inorder to avoid this you can use copy of same list in the iteration.