合并排序函数上的递归令人困惑

Question

I understand the basic principle of a recursion, and it is not confusing at all when dealing with simple things such as calculating a number's factorial, which is the most basic usage of a recursion. However, when you start using Recursion in more complicated environments, it starts getting really really confusing for me.

In my case, I wanted to create a function which used the "merge sort" way to sort the items in a list, using a recursion. So i got to work and this is what i came up with at my first attempt (not really I had to fix some typos):

def merge_sort(ls):
  size = len(ls)
  if size <= 1:
    return ls
  left = merge_sort(ls[:size/2])
  right = merge_sort(ls[size/2:])
  return merge(left, right)

def merge(left, right):
  ls = []
  ln1 = len(left)
  ln2 = len(right)
  length = ln1 + ln2
  while length > 0:
    if not left:
      ls.append(right.pop(0))
      length -= 1
    elif not right:
      ls.append(left.pop(0))
      length -= 1
    else:
      if left[0] < right[0]:
        ls.append(left.pop(0))
        length -= 1
      elif right[0] < left[0]:
        ls.append(right.pop(0))
        length -= 1
  return ls

print merge_sort([8,5,4,6,1])

I clicked run and it worked, but I don't understand why it works. Yes, I created a piece of code which does what was intended to do but I don't understand why.

So "merge_sort" splits the list into two pieces until there is a simple list with a single item. It does this using recursion, and when we get a simple list with a single item it assigns it to a new list, which is left or right . Then i use these 2 new lists in the "merge" function to sort them and merge them together in the right way. So far so good, but what I was expecting when I hit "execute" was for the code to stop here at the first 2 items and print a small list with 2 sorted items. However "merge_sort" keeps going. Even though i believe it should ends after executing:

return merge(left, right)

Now this is my question. Why does it return back to the start of "merge_sort" and how does it save the old values without screwing up the new sorted "small list" generated by "merge" ? I hope my question makes sense to you.

Thank you for the help guys!

Answer 1

the answer to this is a pretty broad one. to fully understand this, you need to understand how a stackbuildup works with recursion and how the pointers in said buildup work.

This is called recursively because the list needs to be broken down by dividing it in half until it is at the smallest possible unit (remainder of 1 or 0) Once this is done, the stack buildup will use its predefined pointers to traceback all of the previous call instances.

this is an indepth programming languages concept.

for more information on stack buildups and call backs reference here. https://en.wikipedia.org/wiki/Call_stack